AI Advances in ICU with an Emphasis on Sepsis Prediction: An Overview
Open accessThe DOI matches its Crossref record; 243 of the 243 references it cites were found in a registry or an index.
Abstract
Artificial intelligence (AI) is increasingly applied in a wide range of healthcare and Intensive Care Unit (ICU) areas to serve—among others—as a tool for disease detection and prediction, as well as for healthcare resources’ management. Since sepsis is a high mortality and rapidly developing organ dysfunction disease afflicting millions in ICUs and costing huge amounts to treat, the area can benefit from the use of AI tools for early and informed diagnosis and antibiotic administration. Additionally, resource allocation plays a crucial role when patient flow is increased, and resources are limited. At the same time, sensitive data use raises the need for ethical guidelines and reflective datasets. Additionally, explainable AI is applied to handle AI opaqueness. This study aims to present existing clinical approaches for infection assessment in terms of scoring systems and diagnostic biomarkers, along with their limitations, and an extensive overview of AI applications in healthcare and ICUs in terms of (a) sepsis detection/prediction and sepsis mortality prediction, (b) length of ICU/hospital stay prediction, and (c) ICU admission/hospitalization prediction after Emergency Department admission, each constituting an important factor towards either prompt interventions and improved patient wellbeing or efficient resource management. Challenges of AI applications in ICU are addressed, along with useful recommendations to mitigate them. Explainable AI applications in ICU are described, and their value in validating, and translating predictions in the clinical setting is highlighted. The most important findings and future directions including multimodal data use and Transformer-based models are discussed. The goal is to make research in AI advances in ICU and particularly sepsis prediction more accessible and provide useful directions on future work.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The third international consensus definitions for sepsis and septic shock (Sepsis-3).
Singer, M., Deutschman, C. S., Seymour, C. W., Shankar-Hari, M., Annane, D., Bauer, M., Bellomo, R., Bernard, G. R., Chiche, J.-D., Coopersmith, C. M., et al. (2016) · JAMA · doi:10.1001/jama.2016.0287 · Cited by 28,250 · cited 1×
“It is also used in the definition of sepsis [97], while its ‘derivative’, namely Quick-SOFA, is used as a screening tool to identify sepsis in patients [84].”
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423 · cited 1×
“Explainability or the area of Explainable Artificial Intelligence (XAI) [3,4] aims to achieve confidence, trustworthiness, accessibility, causality, and transferability in predictions, so that health professionals can understand and correlate the results with the clinical practice [5,6].”
MIMIC-III, a freely accessible critical care database
Johnson, A. E. W., Pollard, T. J., Shen, L., Lehman, L.-W. H., Feng, M., Ghassemi, M., Moody, B., Szolovits, P., Anthony Celi, L., & Mark, R. G. (2016) · Scientific Data · doi:10.1038/sdata.2016.35 · Cited by 8,141 · cited 2×
“Data are getting larger as interoperability increases, more countries and clinical sites share their data [159,188], and dataset(s) are being updated [147,187].”
Le Gall, J.-R. (1993) · JAMA: The Journal of the American Medical Association · doi:10.1001/jama.270.24.2957 · Cited by 7,773 · cited 1×
“SAPS II was validated only in North America and Europe and excluded patients below or at the age of 16 years old, as well as those with burn injuries, coronary artery disease, or a history of cardiac surgery [84,85,88,94,101–103].”
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., & Kang, J. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btz682 · Cited by 7,218 · cited 1×
“To improve model performance, future work is expected to investigate sepsis prediction and other clinical outcomes using multimodal data, Transformer-based models, specific disease cohorts, and be informed and driven by clinical knowledge.”
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01228-7 · Cited by 6,909 · cited 1×
“Model-specific methods (e.g., Gradient-weighted Class Activation Mapping-Grad-CAM [229], Sensitivity Analysis [230], Heatmaps [221,227], Instance-wise variable selection (INVASE) [231], and belief rule-based inference methodology [228]) are applied to ML or DL models with a specific structure or…”
Biomarkers and surrogate endpoints: Preferred definitions and conceptual framework
(2001) · Clinical Pharmacology & Therapeutics · doi:10.1067/mcp.2001.113989 · Cited by 6,762
MIMIC-IV, a freely accessible electronic health record dataset
Johnson, A. E. W., Bulgarelli, L., Shen, L., Gayles, A., Shammout, A., Horng, S., Pollard, T. J., Hao, S., Moody, B., Gow, B., et al. (2023) · Scientific Data · doi:10.1038/s41597-022-01899-x · Cited by 3,401 · cited 2×
“Data are getting larger as interoperability increases, more countries and clinical sites share their data [159,188], and dataset(s) are being updated [147,187].”
Towards A Rigorous Science of Interpretable Machine Learning
Doshi-Velez, F., & Kim, B. (2017) · arXiv · doi:10.48550/arxiv.1702.08608 · Cited by 3,121 · cited 2×
“Explainability or the area of Explainable Artificial Intelligence (XAI) [3,4] aims to achieve confidence, trustworthiness, accessibility, causality, and transferability in predictions, so that health professionals can understand and correlate the results with the clinical practice [5,6].”
The EU General Data Protection Regulation (GDPR)
Voigt, P., & von dem Bussche, A. (2017) · doi:10.1007/978-3-319-57959-7 · Cited by 3,001
Serial Evaluation of the SOFA Score to Predict Outcome in Critically Ill Patients
Ferreira, F. L. (2001) · JAMA · doi:10.1001/jama.286.14.1754 · Cited by 2,862 · cited 2×
“We define sepsis based on the Sepsis-3 definition given in [98].”
Estimation of the Youden Index and its Associated Cutoff Point
Fluss, R., Faraggi, D., & Reiser, B. (2005) · Biometrical Journal · doi:10.1002/bimj.200410135 · Cited by 2,608 · cited 1×
“This method defines the optimal threshold as the point maximizing the Youden function which is the difference between true positive rate (Sensitivity) and false positive rate (1-Specificity) out of all possible thresholds [215].”
IL-6 as a keystone cytokine in health and disease
Hunter, C. A., & Jones, S. A. (2015) · Nature Immunology · doi:10.1038/ni.3153 · Cited by 2,511 · cited 2×
“Interleukin-6 (IL-6) is used as an inflammation biomarker in septic and non-septic patients [74] and as a predictor of disease severity [75].”
Pollard, T. J., Johnson, A. E. W., Raffa, J. D., Celi, L. A., Mark, R. G., & Badawi, O. (2018) · Scientific Data · doi:10.1038/sdata.2018.178 · Cited by 1,993 · cited 2×
“Data are getting larger as interoperability increases, more countries and clinical sites share their data [159,188], and dataset(s) are being updated [147,187].”
Zimmerman, J. E., Kramer, A. A., McNair, D. S., & Malila, F. M. (2006) · Critical Care Medicine · doi:10.1097/01.ccm.0000215112.84523.f0 · Cited by 1,764
Perkins, N. J., & Schisterman, E. F. (2006) · American Journal of Epidemiology · doi:10.1093/aje/kwj063 · Cited by 1,759 · cited 1×
“The threshold is ‘optimal’ when it classifies most of the individuals correctly [212], ensuring maximum sensitivity and specificity.”
Moreno, R. P., Metnitz, P. G. H., Almeida, E., Jordan, B., Bauer, P., Campos, R. A., Iapichino, G., Edbrooke, D., Capuzzo, M., Le Gall, J.-R., et al. (2005) · Intensive Care Medicine · doi:10.1007/s00134-005-2763-5 · Cited by 1,413
Waage, A., Brandtzaeg, P., Halstensen, A., Kierulf, P., & Espevik, T. (1989) · The Journal of experimental medicine · doi:10.1084/jem.169.1.333 · Cited by 1,173 · cited 1×
“In healthy individuals, IL-6 ranges from 0 to 7 pg/mL [76], while IL-6 of more than 1 µ g/mL is indicative of septic shock [76].”
Fisher, A., Rudin, C., & Dominici, F. (2018) · arXiv · doi:10.48550/arxiv.1801.01489 · Cited by 1,165 · cited 1×
“Global methods (e.g., Shapley Additive Explanations-SHAP [222,223], Feature Importance [224]) are used to describe the overall functioning of the model.”
BioGPT: generative pre-trained transformer for biomedical text generation and mining
Luo, R., Sun, L., Xia, Y., Qin, T., Zhang, S., Poon, H., & Liu, T.-Y. (2022) · Briefings in Bioinformatics · doi:10.1093/bib/bbac409 · Cited by 1,150 · cited 1×
“Some popular transformer-based models trained specifically on healthcare data are BioBERT [253], MedBERT [254], BEHRT [255], BioGPT [256], Med-Palm [257], Foresight [258], and Gemini [259].”
The SOFA score—development, utility and challenges of accurate assessment in clinical trials
Lambden, S., Laterre, P. F., Levy, M. M., & Francois, B. (2019) · Critical Care · doi:10.1186/s13054-019-2663-7 · Cited by 979
Multitask learning and benchmarking with clinical time series data
Harutyunyan, H., Khachatrian, H., Kale, D. C., Ver Steeg, G., & Galstyan, A. (2019) · Scientific Data · doi:10.1038/s41597-019-0103-9 · Cited by 958 · cited 4×
“Papers found are split in 3 LOS outcome categories: (i) a continuous outcome in hours/days [149–167] (19, 49%), (ii) a binary outcome of LOS > X days [151,159,168–179] (14, 36%), and (iii) a multiclass outcome of LOS > X days [152,163,180–185] (8, 21%).”
InfluentialRasmy, L., Xiang, Y., Xie, Z., Tao, C., & Zhi, D. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00455-y · Cited by 886 · cited 1×
“Some popular transformer-based models trained specifically on healthcare data are BioBERT [253], MedBERT [254], BEHRT [255], BioGPT [256], Med-Palm [257], Foresight [258], and Gemini [259].”
An interpretable machine learning model for accurate prediction of sepsis in the ICU
Nemati, S., Holder, A., Razmi, F., Stanley, M. D., Clifford, G. D., & Buchman, T. G. (2018) · Critical Care Medicine · doi:10.1097/ccm.0000000000002936 · Cited by 832 · cited 2×
“These methods intend to enable early identification of patients with any type of sepsis, so that clinicians undertake the most appropriate treatment strategy, confidently, enhancing patient prognosis [105].”
Belief rule-base inference methodology using the evidential reasoning Approach-RIMER
Yang, J.-B., Liu, J., Wang, J., Sii, H.-S., & Wang, H.-W. (2006) · IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans · doi:10.1109/tsmca.2005.851270 · Cited by 828 · cited 2×
“Ante-hoc methods (e.g., rule-based [228]) apply explainability during the development and design of the model.”
Defining an Optimal Cut-Point Value in ROC Analysis: An Alternative Approach
Unal, I. (2017) · Computational and Mathematical Methods in Medicine · doi:10.1155/2017/3762651 · Cited by 789 · cited 1×
“There are different ways for identifying the optimal threshold [213].”
Fleuren, L. M., Klausch, T. L. T., Zwager, C. L., Schoonmade, L. J., Guo, T., Roggeveen, L. F., Swart, E. L., Girbes, A. R. J., Thoral, P., Ercole, A., et al. (2020) · Intensive Care Medicine · doi:10.1007/s00134-019-05872-y · Cited by 755
AI in Medical Imaging Informatics: Current Challenges and Future Directions
Panayides, A. S., Amini, A., Filipovic, N. D., Sharma, A., Tsaftaris, S. A., Young, A., Foran, D., Do, N., Golemati, S., Kurc, T., et al. (2020) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.2991043 · Cited by 713 · cited 1×
“This enables the algorithms to predict outcomes for new patients with different clinical characteristics and demographics [2].”
Self-citationPradhan, A. D., Manson, J. E., Rossouw, J. E., Siscovick, D. S., Mouton, C. P., Rifai, N., Wallace, R. B., Jackson, R. D., Pettinger, M. B., & Ridker, P. M. (2002) · JAMA · doi:10.1001/jama.288.8.980 · Cited by 680 · cited 1×
“SOFA was originally used to understand and describe organ disfunction and the complications caused by it in critically ill patients with sepsis [88,96].”
Tuli, S., Basumatary, N., Gill, S. S., Kahani, M., Arya, R. C., Wander, G. S., & Buyya, R. (2020) · Future Generation Computer Systems · doi:10.1016/j.future.2019.10.043 · Cited by 669 · cited 1×
“Regarding cardiovascular diseases, AI has been used to assess the risk of developing a cardiovascular disease [21–27] and identify heart rate severity [28,29].”
Heart Disease Prediction using Machine Learning Techniques
Shah, D., Patel, S., & Bharti, S. K. (2020) · SN Computer Science · doi:10.1007/s42979-020-00365-y · Cited by 658
What do we need to build explainable AI systems for the medical domain?
Holzinger, A., Biemann, C., Pattichis, C. S., & Kell, D. B. (2017) · arXiv · doi:10.48550/arxiv.1712.09923 · Cited by 636 · cited 1×
“Explainability or the area of Explainable Artificial Intelligence (XAI) [3,4] aims to achieve confidence, trustworthiness, accessibility, causality, and transferability in predictions, so that health professionals can understand and correlate the results with the clinical practice [5,6].”
Self-citationHaq, A. U., Li, J. P., Memon, M. H., Nazir, S., & Sun, R. (2018) · Mobile Information Systems · doi:10.1155/2018/3860146 · Cited by 621
Sensitivity Analysis as an Ingredient of Modeling
Saltelli, A., Tarantola, S., & Campolongo, F. (2000) · Cited by 604
BEHRT: Transformer for Electronic Health Records
Li, Y., Rao, S., Solares, J. R. A., Hassaine, A., Ramakrishnan, R., Canoy, D., Zhu, Y., Rahimi, K., & Salimi-Khorshidi, G. (2020) · Scientific Reports · doi:10.1038/s41598-020-62922-y · Cited by 595 · cited 1×
“Some popular transformer-based models trained specifically on healthcare data are BioBERT [253], MedBERT [254], BEHRT [255], BioGPT [256], Med-Palm [257], Foresight [258], and Gemini [259].”
Effective Heart Disease Prediction Using Machine Learning Techniques
Bhatt, C. M., Patel, P., Ghetia, T., & Mazzeo, P. L. (2023) · Algorithms · doi:10.3390/a16020088 · Cited by 586
Hou, N., Li, M., He, L., Xie, B., Wang, L., Zhang, R., Yu, Y., Sun, X., Pan, Z., & Wang, K. (2020) · Journal of Translational Medicine · doi:10.1186/s12967-020-02620-5 · Cited by 568 · cited 7×
“For sepsis mortality prediction models, built-in feature importance [113,240–242], SHAP [240,241,243–246], LIME [127,245,247], Break Down [241], Ceteris Paribus [241], and INVASE [242] are the explainable AI techniques used in literature.”
InfluentialHL7 FHIR: An Agile and RESTful approach to healthcare information exchange
Bender, D., & Sartipi, K. (2013) · Proceedings of the 26th IEEE International Symposium on Computer-Based Medical Systems · doi:10.1109/cbms.2013.6627810 · Cited by 540 · cited 1×
“In order to allow for more data to be collected efficiently, standardization procedures and interoperability profiles like ICD [209] and FIHR [210] are essential.”
Emergency department triage prediction of clinical outcomes using machine learning models
Raita, Y., Goto, T., Faridi, M. K., Brown, D. F. M., Camargo, C. A., & Hasegawa, K. (2019) · Critical Care · doi:10.1186/s13054-019-2351-7 · Cited by 519 · cited 6×
“Some studies suggest that ML and DL models perform similarly [191,192].”
InfluentialC-Reactive Protein Levels Correlate With Mortality and Organ Failure in Critically Ill Patientsa
Lobo, S. M. A., Lobo, F. R. M., Bota, D. P., Lopes-Ferreira, F., Soliman, H. M., Meélot, C., & Vincent, J.-L. (2003) · Chest · doi:10.1378/chest.123.6.2043 · Cited by 448 · cited 1×
“It is associated with ICU-acquired infection [51], hospital readmission in patients with heart failure [60], ICU readmission, unexpected mortality after ICU discharge [61–63], as well as ICU mortality [55] and ICU mortality of COVID-19 patients [67].”
A., S., & R., S. (2023) · Decision Analytics Journal · doi:10.1016/j.dajour.2023.100230 · Cited by 432 · cited 6×
“Healthcare professionals’ input in features included in models should also be prioritized.”
InfluentialGoh, K. H., Wang, L., Yeow, A. Y. K., Poh, H., Li, K., Yeow, J. J. L., & Tan, G. Y. H. (2021) · Nature Communications · doi:10.1038/s41467-021-20910-4 · Cited by 415 · cited 3×
“[112,115,118,123,126–128, 130– 137, 139, In summary, we differentiate between models used for detecting sepsis presence versus the prediction of sepsis at a future time.”
Machine Learning Algorithms For Breast Cancer Prediction And Diagnosis
Naji, M. A., Filali, S. E., Aarika, K., Benlahmar, E. H., Abdelouhahid, R. A., & Debauche, O. (2021) · Procedia Computer Science · doi:10.1016/j.procs.2021.07.062 · Cited by 385 · cited 1×
“Moreover, studies use AI for predicting a diverse set of pathologies, such as the diagnosis of ovarian tumors [8] and the survival of patients with ovarian cancer [9], early breast cancer prediction and diagnosis [10,11], dermatological cancer recognition [12], and lung cancer diagnosis [13].”
Early Prediction of Sepsis From Clinical Data: The PhysioNet/Computing in Cardiology Challenge 2019
Reyna, M. A., Josef, C. S., Jeter, R., Shashikumar, S. P., Westover, M. B., Nemati, S., Clifford, G. D., & Sharma, A. (2020) · Critical Care Medicine · doi:10.1097/ccm.0000000000004145 · Cited by 383
Classification and prediction of diabetes disease using machine learning paradigm
Maniruzzaman, M., Rahman, M. J., Ahammed, B., & Abedin, M. M. (2020) · Health Information Science and Systems · doi:10.1007/s13755-019-0095-z · Cited by 369 · cited 1×
“Studies have further applied AI for the classification of patients into diabetic or non-diabetic [15,16], the risk prediction of developing type 2 diabetes [17,18], diabetes diagnosis [19], and blood glucose level predictions [20].”
Woldaregay, A. Z., Årsand, E., Walderhaug, S., Albers, D., Mamykina, L., Botsis, T., & Hartvigsen, G. (2019) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2019.07.007 · Cited by 358 · cited 1×
“Studies have further applied AI for the classification of patients into diabetic or non-diabetic [15,16], the risk prediction of developing type 2 diabetes [17,18], diabetes diagnosis [19], and blood glucose level predictions [20].”
Machine learning based early warning system enables accurate mortality risk prediction for COVID-19
Gao, Y., Cai, G.-Y., Fang, W., Li, H.-Y., Wang, S.-Y., Chen, L., Yu, Y., Liu, D., Xu, S., Cui, P.-F., et al. (2020) · Nature Communications · doi:10.1038/s41467-020-18684-2 · Cited by 349 · cited 1×
“ML techniques have been used to predict whether patients are infected by COVID-19 [30], identify its spread [31,32], predict the number of discharged patients and deaths [33], and predict COVID-19 mortality [34,35] to prioritize triage and hospitalization [35].”
Explanatory Model Analysis: Explore, Explain and Examine Predictive Models
Biecek, P., & Burzykowski, T. (2021) · Chapman & Hall/CRC eBooks · doi:10.1201/9780429027192 · Cited by 344
Higgins, T. L., Teres, D., Copes, W. S., Nathanson, B. H., Stark, M., & Kramer, A. A. (2007) · Critical Care Medicine · doi:10.1097/01.ccm.0000257337.63529.9f · Cited by 333 · cited 3×
“Early identification and immediate intervention are keys to sepsis treatment; while scoring systems and diagnostic biomarkers can be insufficient to detect or predict the response to infection and are accompanied by limitations (see Table 2; [48,104,106–109]).”
Pourhomayoun, M., & Shakibi, M. (2021) · Smart Health · doi:10.1016/j.smhl.2020.100178 · Cited by 331 · cited 1×
“ML techniques have been used to predict whether patients are infected by COVID-19 [30], identify its spread [31,32], predict the number of discharged patients and deaths [33], and predict COVID-19 mortality [34,35] to prioritize triage and hospitalization [35].”
Solid phase radioimmunoassays for human C-reactive protein
Shine, B., de Beer, F. C., & Pepys, M. B. (1981) · Clinica Chimica Acta · doi:10.1016/0009-8981(81)90005-x · Cited by 319
Predictive modelling and analytics for diabetes using a machine learning approach
Kaur, H., & Kumari, V. (2022) · Applied Computing and Informatics · doi:10.1016/j.aci.2018.12.004 · Cited by 317
The legal and ethical concerns that arise from using complex predictive analytics in health care
Cohen, I. G., Amarasingham, R., Shah, A., Xie, B., & Lo, B. (2014) · Health Affairs · doi:10.1377/hlthaff.2014.0048 · Cited by 300 · cited 1×
“With the rapid development of AI, the discussion on ethics shifted towards the ethical implications of using ML methods for prognosis [217,218].”
MIMIC-Extract: a data extraction, preprocessing, and representation pipeline for MIMIC-III
Wang, S., McDermott, M. B. A., Chauhan, G., Ghassemi, M., Hughes, M. C., & Naumann, T. (2020) · ACM Conference on Health, Inference, and Learning (CHIL) · doi:10.1145/3368555.3384469 · Cited by 290
Goto, T., Camargo, C. A., Faridi, M. K., Freishtat, R. J., & Hasegawa, K. (2019) · JAMA Network Open · doi:10.1001/jamanetworkopen.2018.6937 · Cited by 285 · cited 5×
“Some studies suggest that ML and DL models perform similarly [191,192].”
InfluentialImplementation of Machine Learning Model to Predict Heart Failure Disease
Alotaibi, F. S. (2019) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2019.0100637 · Cited by 278 · cited 1×
“Regarding cardiovascular diseases, AI has been used to assess the risk of developing a cardiovascular disease [21–27] and identify heart rate severity [28,29].”
Computer aided lung cancer diagnosis with deep learning algorithms
Sun, W., Zheng, B., & Qian, W. (2016) · SPIE Proceedings · doi:10.1117/12.2216307 · Cited by 277
Chaki, J., Thillai Ganesh, S., Cidham, S. K., & Ananda Theertan, S. (2022) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2020.06.013 · Cited by 270 · cited 1×
“Additionally, AI methods are cost-effective for reducing ophthalmic complications and preventable blindness associated with diabetes [14].”
Mortality predictions in the intensive care unit: Comparing physicians with scoring systems*
Sinuff, T., Adhikari, N. K. J., Cook, D. J., Schünemann, H. J., Griffith, L. E., Rocker, G., & Walter, S. D. (2006) · Critical Care Medicine · doi:10.1097/01.ccm.0000201881.58644.41 · Cited by 262 · cited 1×
“Early identification and immediate intervention are keys to sepsis treatment; while scoring systems and diagnostic biomarkers can be insufficient to detect or predict the response to infection and are accompanied by limitations (see Table 2; [48,104,106–109]).”
Prediction of Type 2 Diabetes Based on Machine Learning Algorithm
Deberneh, H. M., & Kim, I. (2021) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph18063317 · Cited by 261 · cited 1×
“Studies have further applied AI for the classification of patients into diabetic or non-diabetic [15,16], the risk prediction of developing type 2 diabetes [17,18], diabetes diagnosis [19], and blood glucose level predictions [20].”
Contributions to the Theory of Games (AM-28), Volume II
(1953) · Princeton University Press eBooks · doi:10.1515/9781400881970 · Cited by 260
C-Reactive Protein as a Predictor of Improvement and Readmission in Heart Failure
Alonso-Martínez, J. L., Llorente-Diez, B., Echegaray-Agara, M., Olaz-Preciado, F., Urbieta-Echezarreta, M., & González-Arencibia, C. (2002) · European Journal of Heart Failure · doi:10.1016/s1388-9842(02)00021-1 · Cited by 244 · cited 1×
“It is associated with ICU-acquired infection [51], hospital readmission in patients with heart failure [60], ICU readmission, unexpected mortality after ICU discharge [61–63], as well as ICU mortality [55] and ICU mortality of COVID-19 patients [67].”
Haque, A. B., Islam, A. K. M. N., & Mikalef, P. (2023) · Technological Forecasting and Social Change · doi:10.1016/j.techfore.2022.122120 · Cited by 240 · cited 1×
“Explainability or the area of Explainable Artificial Intelligence (XAI) [3,4] aims to achieve confidence, trustworthiness, accessibility, causality, and transferability in predictions, so that health professionals can understand and correlate the results with the clinical practice [5,6].”
Scoring systems in the intensive care unit: A compendium
Rapsang, A., & Shyam, D. C. (2014) · Indian Journal of Critical Care Medicine · doi:10.4103/0972-5229.130573 · Cited by 234 · cited 1×
“An overview of the selection process of the studies is depicted in Appendix A Figure A1.”
Large AI Models in Health Informatics: Applications, Challenges, and the Future
Qiu, J., Li, L., Sun, J., Peng, J., Shi, P., Zhang, R., Dong, Y., Lam, K., Lo, F. P.-W., Xiao, B., et al. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3316750 · Cited by 233 · cited 1×
“Additionally, there is strong interest in using Generative AI [251,252] to generate data for rare disease patients or for cases where text, image, and/or video data are limited.”
Early Prediction of Sepsis in the ICU Using Machine Learning: A Systematic Review
Moor, M., Rieck, B., Horn, M., Jutzeler, C. R., & Borgwardt, K. (2021) · Frontiers in Medicine · doi:10.3389/fmed.2021.607952 · Cited by 231
Ljungström, L., Pernestig, A.-K., Jacobsson, G., Andersson, R., Usener, B., & Tilevik, D. (2017) · PLOS ONE · doi:10.1371/journal.pone.0181704 · Cited by 219 · cited 2×
“…bacterial infection: >2 ng/mL [80] - In-hospital mortality predictor of COVID-19 patients: ≥ 0.56 ng/mL on ICU admission [73] WBC - Biomarker of infection [48] - Immature granulocytes [81] - NLR [82,83] - Non-infectious mimics, drugs, and comorbidities [48] Refer to Glossary section for acronyms.”
InfluentialEscarce, J. J. (1990) · JAMA: The Journal of the American Medical Association · doi:10.1001/jama.1990.03450180053028 · Cited by 213
Severity scoring systems in the critically ill
Bouch, D. C., & Thompson, J. P. (2008) · Continuing Education in Anaesthesia Critical Care & Pain · doi:10.1093/bjaceaccp/mkn033 · Cited by 205
Melanoma Skin Cancer Detection using Image Processing and Machine Learning
M, V. M. (2019) · International Journal of Trend in Scientific Research and Development · doi:10.31142/ijtsrd23936 · Cited by 196 · cited 1×
“Moreover, studies use AI for predicting a diverse set of pathologies, such as the diagnosis of ovarian tumors [8] and the survival of patients with ovarian cancer [9], early breast cancer prediction and diagnosis [10,11], dermatological cancer recognition [12], and lung cancer diagnosis [13].”
Henriquez-Camacho, C., & Losa, J. (2014) · BioMed Research International · doi:10.1155/2014/547818 · Cited by 194
Amini, M. M., Jesus, M., Sheikholeslami, D. F., Alves, P. J. P., Benam, A. H., & Hariri, F. (2023) · Machine Learning and Knowledge Extraction · doi:10.3390/make5030053 · Cited by 192 · cited 1×
“With the rapid development of AI, the discussion on ethics shifted towards the ethical implications of using ML methods for prognosis [217,218].”
Early detection of sepsis utilizing deep learning on electronic health record event sequences
Lauritsen, S. M., Kalør, M. E., Kongsgaard, E. L., Lauritsen, K. M., Jørgensen, M. J., Lange, J., & Thiesson, B. (2020) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2020.101820 · Cited by 190
A Machine Learning Model for Accurate Prediction of Sepsis in ICU Patients
Wang, D., Li, J., Sun, Y., Ding, X., Zhang, X., Liu, S., Han, B., Wang, H., Duan, X., & Sun, T. (2021) · Frontiers in Public Health · doi:10.3389/fpubh.2021.754348 · Cited by 186 · cited 4×
“The most important features according to built-in model importance are usually vital signs (e.g., heart rate, respiratory rate) [112,123,135,136,141] and laboratory values (e.g., Platelets, lactate) [112,113,117,119,124,126,141,142,145].”
InfluentialBuilding Risk Prediction Models for Type 2 Diabetes Using Machine Learning Techniques
Xie, Z., Nikolayeva, O., Luo, J., & Li, D. (2019) · Preventing Chronic Disease · doi:10.5888/pcd16.190109 · Cited by 186 · cited 1×
“Studies have further applied AI for the classification of patients into diabetic or non-diabetic [15,16], the risk prediction of developing type 2 diabetes [17,18], diabetes diagnosis [19], and blood glucose level predictions [20].”
Using machine learning methods to predict in-hospital mortality of sepsis patients in the ICU
Kong, G., Lin, K., & Hu, Y. (2020) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-020-01271-2 · Cited by 180 · cited 1×
“Most of the ML models are tree-based ones like Extreme Gradient Boosting (XGB) [113,116,120,124,125,141], Random Forest (RF) [117,129,145], Gradient Boosting (GB) [114], Light Gradient Boosting Machine (LGBM) [119], and AdaBoost [144].”
Using Data Mining to Predict Hospital Admissions From the Emergency Department
Graham, B., Bond, R., Quinn, M. A., & Mulvenna, M. D. (2018) · IEEE Access · doi:10.1109/access.2018.2808843 · Cited by 173 · cited 1×
“As expected, for sepsis prediction, several studies suggest that a longer observation window [133] and a shorter prediction window [133,143] increase model performance.”
Variability in outcome and resource use in intensive care units
Rothen, H. U., Stricker, K., Einfalt, J., Bauer, P., Metnitz, P. G. H., Moreno, R. P., & Takala, J. (2007) · Intensive Care Medicine · doi:10.1007/s00134-007-0690-3 · Cited by 172 · cited 1×
“SAPS II was validated only in North America and Europe and excluded patients below or at the age of 16 years old, as well as those with burn injuries, coronary artery disease, or a history of cardiac surgery [84,85,88,94,101–103].”
Yuan, K.-C., Tsai, L.-W., Lee, K.-H., Cheng, Y.-W., Hsu, S.-C., Lo, Y.-S., & Chen, R.-J. (2020) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2020.104176 · Cited by 170 · cited 4×
“[116,124,141,142,145] of them used a ML model with AUC ranging from 0.86 to 0.97, while just 1 study [122] uses a DL model with an AUC of 0.835.”
InfluentialComparing machine learning algorithms for predicting ICU admission and mortality in COVID-19
Subudhi, S., Verma, A., Patel, A. B., Hardin, C. C., Khandekar, M. J., Lee, H., McEvoy, D., Stylianopoulos, T., Munn, L. L., Dutta, S., et al. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00456-x · Cited by 170 · cited 2×
“Other important features are previous visit outcomes [198], ambulance use [192], intravenous therapy (IVT) [193], and different lab values [203,204,208].”
Kraljevic, Z., Bean, D., Shek, A., Bendayan, R., Hemingway, H., Yeung, J. A., Deng, A., Balston, A., Ross, J., Idowu, E., et al. (2024) · The Lancet Digital Health · doi:10.1016/s2589-7500(24)00025-6 · Cited by 169 · cited 1×
“Some popular transformer-based models trained specifically on healthcare data are BioBERT [253], MedBERT [254], BEHRT [255], BioGPT [256], Med-Palm [257], Foresight [258], and Gemini [259].”
Predicting sepsis with a recurrent neural network using the MIMIC III database
Scherpf, M., Gräßer, F., Malberg, H., & Zaunseder, S. (2019) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2019.103395 · Cited by 167 · cited 2×
“As expected, for sepsis prediction, several studies suggest that a longer observation window [133] and a shorter prediction window [133,143] increase model performance.”
Schmit, X., & Vincent, J. L. (2008) · Infection · doi:10.1007/s15010-007-7077-9 · Cited by 165 · cited 1×
“It does not consider chronic illnesses [84–86, SAPS assesses mortality risk and can be used to compare resources amongst ICUs.”
Hu, C., Li, L., Huang, W., Wu, T., Xu, Q., Liu, J., & Hu, B. (2022) · Infectious Diseases and Therapy · doi:10.1007/s40121-022-00628-6 · Cited by 160
Artificial Intelligence in Ovarian Cancer Diagnosis
Akazawa, M., & Hashimoto, K. (2020) · Anticancer Research · doi:10.21873/anticanres.14482 · Cited by 157 · cited 1×
“Moreover, studies use AI for predicting a diverse set of pathologies, such as the diagnosis of ovarian tumors [8] and the survival of patients with ovarian cancer [9], early breast cancer prediction and diagnosis [10,11], dermatological cancer recognition [12], and lung cancer diagnosis [13].”
An explainable machine learning framework for lung cancer hospital length of stay prediction
Alsinglawi, B., Alshari, O., Alorjani, M., Mubin, O., Alnajjar, F., Novoa, M., & Darwish, O. (2022) · Scientific Reports · doi:10.1038/s41598-021-04608-7 · Cited by 153 · cited 5×
“According to this study, a prespecified value is sometimes used as a threshold [169], while in other cases it is identified based on either the mean [170], median [151,168], 75th percentile [171,172] of the study population outcome, previous studies [173], or clinical importance [174].”
InfluentialIslam, K. R., Prithula, J., Kumar, J., Tan, T. L., Reaz, M. B. I., Sumon, M. S. I., & Chowdhury, M. E. H. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12175658 · Cited by 153 · cited 1×
“While similar review studies exist, they focus on a single ICU outcome [36–43], explore studies that use a single type of data [36,38,42], do not elaborate on clinical approaches used in ICU [36–44], and/or do not significantly guide towards future work [36,37,43].”
DeepAISE – An interpretable and recurrent neural survival model for early prediction of sepsis
Shashikumar, S. P., Josef, C. S., Sharma, A., & Nemati, S. (2021) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2021.102036 · Cited by 146 · cited 1×
“Sensitivity analysis [232,237], LIME [127,238], heatmaps [239] and Grad-CAM [234] are also used.”
Reny, J., Vuagnat, A., Ract, C., Benoit, M., Safar, M. E., & Fagon, J.-Y. (2002) · Critical Care Medicine · doi:10.1097/00003246-200203000-00006 · Cited by 143
Póvoa, P. M. S. R., Coelho, L., Almeida, E., Fernandes, A., Mealha, R., Moreira, P., & Sabino, H. A. C. (2006) · Critical Care · doi:10.1186/cc4892 · Cited by 143 · cited 1×
“…[ C-Reactive Protein (CRP) is used as a biomarker for Systemic Inflammatory Response Syndrome (SIRS), infection and sepsis [49], for the diagnosis of neonatal sepsis [50], and for the differential diagnosis of bacterial versus viral infections [56,57] and their early identification [58,59].”
Ho, K. M., Lee, K. Y., DOBB, G. J., & Webb, S. A. R. (2007) · Intensive Care Medicine · doi:10.1007/s00134-007-0928-0 · Cited by 139 · cited 1×
“It is associated with ICU-acquired infection [51], hospital readmission in patients with heart failure [60], ICU readmission, unexpected mortality after ICU discharge [61–63], as well as ICU mortality [55] and ICU mortality of COVID-19 patients [67].”
Applying Best Machine Learning Algorithms for Breast Cancer Prediction and Classification
Khourdifi, Y., & Bahaj, M. (2018) · IEEE International Conference on Electronics, Control, Optimization and Computer Science (ICECOCS) · doi:10.1109/icecocs.2018.8610632 · Cited by 136 · cited 1×
“Moreover, studies use AI for predicting a diverse set of pathologies, such as the diagnosis of ovarian tumors [8] and the survival of patients with ovarian cancer [9], early breast cancer prediction and diagnosis [10,11], dermatological cancer recognition [12], and lung cancer diagnosis [13].”
Drożdż, K., Nabrdalik, K., Kwiendacz, H., Hendel, M., Olejarz, A., Tomasik, A. R., Bartman, W., Nalepa, J., Gumprecht, J., & Lip, G. Y. H. (2022) · Cardiovascular Diabetology · doi:10.1186/s12933-022-01672-9 · Cited by 136
Prediction algorithm for ICU mortality and length of stay using machine learning
Iwase, S., Nakada, T., Shimada, T., Oami, T., Shimazui, T., Takahashi, N., Yamabe, J., Yamao, Y., & Kawakami, E. (2022) · Scientific Reports · doi:10.1038/s41598-022-17091-5 · Cited by 133 · cited 5×
“According to this study, a prespecified value is sometimes used as a threshold [169], while in other cases it is identified based on either the mean [170], median [151,168], 75th percentile [171,172] of the study population outcome, previous studies [173], or clinical importance [174].”
InfluentialLilly, C. M., Zuckerman, I. H., Badawi, O., & Riker, R. R. (2011) · CHEST Journal · doi:10.1378/chest.11-0718 · Cited by 132
Breast Cancer Detection in the IOT Health Environment Using Modified Recursive Feature Selection
Memon, M. H., Li, J., Haq, A. U., Memon, M. H., & Zhou, W. (2019) · Wireless Communications and Mobile Computing · doi:10.1155/2019/5176705 · Cited by 130 · cited 1×
“AI is increasingly used in healthcare in the investigation of sicknesses that are hard to diagnose [7], including cancer, diabetes, cardiovascular diseases, COVID-19, etc., by identifying the most important factors for patients’ risk prediction and/or diagnosis.”
Xie, F., Zhou, J. R., Lee, J. W., Tan, M., Li, S., Rajnthern, L. S. O., Chee, M. L., Chakraborty, B., Wong, A.-K. I., Dagan, A. S., et al. (2022) · Scientific Data · doi:10.1038/s41597-022-01782-9 · Cited by 129 · cited 1×
“ML techniques have emerged as effective for ED triage prediction of hospitalization models, achieving high accuracy [190,191].”
Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”
Shashikumar, S. P., Wardi, G., Malhotra, A., & Nemati, S. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00504-6 · Cited by 125 · cited 5×
“The most important features according to built-in model importance are usually vital signs (e.g., heart rate, respiratory rate) [112,123,135,136,141] and laboratory values (e.g., Platelets, lactate) [112,113,117,119,124,126,141,142,145].”
InfluentialLiSep LSTM: A Machine Learning Algorithm for Early Detection of Septic Shock
Fagerström, J., Bång, M., Wilhelms, D., & Chew, M. S. (2019) · Scientific Reports · doi:10.1038/s41598-019-51219-4 · Cited by 124 · cited 2×
“As expected, for sepsis prediction, several studies suggest that a longer observation window [133] and a shorter prediction window [133,143] increase model performance.”
Yan, M. Y., Gustad, L. T., & Nytrø, Ø. (2022) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocab236 · Cited by 119 · cited 1×
“While similar review studies exist, they focus on a single ICU outcome [36–43], explore studies that use a single type of data [36,38,42], do not elaborate on clinical approaches used in ICU [36–44], and/or do not significantly guide towards future work [36,37,43].”
Alghatani, K., Ammar, N., Rezgui, A., & Shaban‐Nejad, A. (2021) · JMIR Medical Informatics · doi:10.2196/21347 · Cited by 113 · cited 6×
“According to this study, a prespecified value is sometimes used as a threshold [169], while in other cases it is identified based on either the mean [170], median [151,168], 75th percentile [171,172] of the study population outcome, previous studies [173], or clinical importance [174].”
InfluentialMachine learning for early detection of sepsis: an internal and temporal validation study
Bedoya, A. D., Futoma, J., Clement, M. E., Corey, K., Brajer, N., Lin, A., Simons, M. G., Gao, M., Nichols, M., Balu, S., et al. (2020) · JAMIA Open · doi:10.1093/jamiaopen/ooaa006 · Cited by 109 · cited 3×
“DL models proposed are usually temporal ones, like the Long Short Term Memory (LSTM) [122,127,135,143], the Recurrent Neural Network (RNN) [131,133], the Convolutional Neural Network (CNN) [115,140], the Temporal Convolutional Network (TCN) [118], or combinations of them [132,134].”
Mortality Probability Model III and Simplified Acute Physiology Score II
Vasilevskis, E. E., Kuzniewicz, M. W., Cason, B. A., Lane, R. K., Dean, M. L., Clay, T., Rennie, D. J., Vittinghoff, E., & Dudley, R. A. (2009) · CHEST Journal · doi:10.1378/chest.08-2591 · Cited by 108
Haniffa, R., Isaam, I., Silva, A. P. D., Dondorp, A. M., & de Keizer, N. F. (2018) · Critical Care · doi:10.1186/s13054-017-1930-8 · Cited by 107 · cited 2×
“Later it had been validated for use in critically ill patients with non-sepsis related organ disfunction and as a method for predicting mortality rate [88,96].”
Comparing different feature selection algorithms for cardiovascular disease prediction
Hasan, N., & Bao, Y. (2020) · Health and Technology · doi:10.1007/s12553-020-00499-2 · Cited by 106
Automated prediction of sepsis using temporal convolutional network
Kok, C., Jahmunah, V., Oh, S. L., Zhou, X., Gururajan, R., Tao, X., Cheong, K. H., Gururajan, R., Molinari, F., & Acharya, U. R. (2020) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2020.103957 · Cited by 105 · cited 3×
“%) prediction window used is 6 h [118–120 , 138–”
Countering cognitive biases in minimising low value care
Scott, I. A., Soon, J., Elshaug, A. G., & Lindner, R. (2017) · Medical Journal of Australia · doi:10.5694/mja16.00999 · Cited by 101
Su, L., Zheng, X., Chang, F., Ma, Y., Liu, S., Jiang, H., Wang, H., Li, D., Chen, H., Zhou, X. Q., et al. (2021) · Frontiers in Medicine · doi:10.3389/fmed.2021.664966 · Cited by 100 · cited 5×
“According to this study, a prespecified value is sometimes used as a threshold [169], while in other cases it is identified based on either the mean [170], median [151,168], 75th percentile [171,172] of the study population outcome, previous studies [173], or clinical importance [174].”
InfluentialMilenković, M., Hadžibegović, A., Kovac, M. K., Jovanović, B., Stanisavljević, J., Djikic, M., Šijan, D., Ladjevic, N. G., Palibrk, I., Djukanović, M., et al. (2022) · Oxidative Medicine and Cellular Longevity · doi:10.1155/2022/8997709 · Cited by 99 · cited 3×
“For COVID-19 patients, IL-6 of ≥ 74.98 pg/mL on ICU admission is a predictor of in-hospital mortality [73] (see Figure 1).”
InfluentialAn Update on C-reactive Protein for Intensivists
Ho, K. M., & Lipman, J. (2009) · Anaesthesia and Intensive Care · doi:10.1177/0310057x0903700217 · Cited by 98 · cited 2×
“CRP can be used as a valuable tool to monitor progress as it responds to therapy against inflammation [54].”
An interpretable deep-learning model for early prediction of sepsis in the emergency department
Zhang, D., Yin, C., Hunold, K. M., Jiang, X., Caterino, J. M., & Zhang, P. (2021) · Patterns · doi:10.1016/j.patter.2020.100196 · Cited by 98
Benchmarking machine learning models on multi-centre eICU critical care dataset
Sheikhalishahi, S., Balaraman, V., & Osmani, V. (2020) · PLOS ONE · doi:10.1371/journal.pone.0235424 · Cited by 89 · cited 2×
“%) studies [153,155,156,169].”
Caution when using prognostic models: A prospective comparison of 3 recent prognostic models
Nassar, A. P., Mocelin, A. O., Nunes, A. L. B., Giannini, F. P., Brauer, L., Andrade, F. M., & Dias, C. A. (2011) · Journal of Critical Care · doi:10.1016/j.jcrc.2011.08.016 · Cited by 88
INVASE: Instance-wise Variable Selection using Neural Networks
Yoon, J., Jordon, J., & van der Schaar, M. (2018) · International Conference on Learning Representations · Cited by 88
Erez, A., Shental, O., Tchebiner, J. Z., Laufer‐Perl, M., Wasserman, A., Sella, T., & Guzner‐Gur, H. (2014) · PubMed · Cited by 88
Jiang, Z., Bo, L., Xu, Z., Song, Y., Wang, J., Wen, P., Wan, X., Yang, T., Deng, X., & Bian, J. (2021) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2021.106040 · Cited by 86 · cited 1×
“For sepsis mortality prediction models, built-in feature importance [113,240–242], SHAP [240,241,243–246], LIME [127,245,247], Break Down [241], Ceteris Paribus [241], and INVASE [242] are the explainable AI techniques used in literature.”
Forecast and prediction of COVID-19 using machine learning
Painuli, D., Mishra, D., Bhardwaj, S., & Aggarwal, M. (2021) · Elsevier eBooks · doi:10.1016/b978-0-12-824536-1.00027-7 · Cited by 85 · cited 1×
“ML techniques have been used to predict whether patients are infected by COVID-19 [30], identify its spread [31,32], predict the number of discharged patients and deaths [33], and predict COVID-19 mortality [34,35] to prioritize triage and hospitalization [35].”
Predicting Intensive Care Unit Outcome with Scoring Systems: Underlying Concepts and Principles
Kollef, M. H., & Schuster, D. P. (1994) · Critical Care Clinics · doi:10.1016/s0749-0704(18)30141-6 · Cited by 83 · cited 1×
“They are also used to evaluate therapeutic interventions in patients with acute respiratory distress syndrome or sepsis [86,87], and for benchmarking of ICU performance and improvement of quality of care [84].”
MGP-AttTCN: An interpretable machine learning model for the prediction of sepsis
Rosnati, M., & Fortuin, V. (2021) · PLoS ONE · doi:10.1371/journal.pone.0251248 · Cited by 81 · cited 1×
“Sensitivity analysis [232,237], LIME [127,238], heatmaps [239] and Grad-CAM [234] are also used.”
Application of machine learning techniques for predicting survival in ovarian cancer
Sorayaie Azar, A., Babaei Rikan, S., Naemi, A., Bagherzadeh Mohasefi, J., Pirnejad, H., Bagherzadeh Mohasefi, M., & Wiil, U. K. (2022) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-022-02087-y · Cited by 81 · cited 1×
“Moreover, studies use AI for predicting a diverse set of pathologies, such as the diagnosis of ovarian tumors [8] and the survival of patients with ovarian cancer [9], early breast cancer prediction and diagnosis [10,11], dermatological cancer recognition [12], and lung cancer diagnosis [13].”
Predictive analytics for hospital admissions from the emergency department using triage information
Araz, Ö. M., Olson, D., & Ramírez–Nafarrate, A. (2018) · International Journal of Production Economics · doi:10.1016/j.ijpe.2018.11.024 · Cited by 79 · cited 1×
“[195] found that performance did improve by using a larger dataset.”
An evaluation of time series summary statistics as features for clinical prediction tasks
Guo, C., Lu, M., & Chen, J. (2020) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-020-1063-x · Cited by 78 · cited 3×
“Papers found are split in 3 LOS outcome categories: (i) a continuous outcome in hours/days [149–167] (19, 49%), (ii) a binary outcome of LOS > X days [151,159,168–179] (14, 36%), and (iii) a multiclass outcome of LOS > X days [152,163,180–185] (8, 21%).”
Yang, A.-P., Liu, J., Yue, L.-H., Wang, H.-Q., Yang, W., & Yang, G. (2015) · Clinical Chemistry and Laboratory Medicine (CCLM) · doi:10.1515/cclm-2015-0277 · Cited by 75 · cited 3×
“PCT - PCT test for early identification of infections [58,59] -”
Abbas, A., Mosseri, J., Lex, J. R., Toor, J., Ravi, B., Khalil, E. B., & Whyne, C. M. (2021) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2021.104670 · Cited by 74
Moor, M., Horn, M., Rieck, B., Roqueiro, D., & Borgwardt, K. M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.01659 · Cited by 71
Saadatmand, S., Salimifard, K., Mohammadi, R., Kuiper, A., Marzban, M., & Farhadi, A. (2022) · Annals of Operations Research · doi:10.1007/s10479-022-04984-x · Cited by 68 · cited 3×
“[171,175] use a DL model with AUC of 0.76 and 0.915, respectively, while 12 [151,159,168–170 , 172– 174”
InfluentialEarly Prediction of Sepsis Based on Machine Learning Algorithm
Zhao, X., Shen, W., & Wang, G. (2021) · Computational Intelligence and Neuroscience · doi:10.1155/2021/6522633 · Cited by 67 · cited 3×
“The most important features according to built-in model importance are usually vital signs (e.g., heart rate, respiratory rate) [112,123,135,136,141] and laboratory values (e.g., Platelets, lactate) [112,113,117,119,124,126,141,142,145].”
What’s new in ICU in 2050: big data and machine learning
Bailly, S., Meyfroidt, G., & Timsit, J. (2017) · Intensive Care Medicine · doi:10.1007/s00134-017-5034-3 · Cited by 66 · cited 1×
“Early identification and immediate intervention are keys to sepsis treatment; while scoring systems and diagnostic biomarkers can be insufficient to detect or predict the response to infection and are accompanied by limitations (see Table 2; [48,104,106–109]).”
De Hond, A., Raven, W., Schinkelshoek, L., Gaakeer, M., Ter Avest, E., Sir, O., Lameijer, H., Hessels, R. A., Reijnen, R., De Jonge, E., et al. (2021) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2021.104496 · Cited by 66 · cited 1×
“%) make predictions at triage [191–205], whereas some make predictions at 30 min [206], 1 h [198], 2 h [206,207] and hourly [208] after patient arrival at ED.”
Xiong, X.-L., Zhang, R.-X., Bi, Y., Zhou, W., Yu, Y., & Zhu, D. (2019) · Current Medical Science · doi:10.1007/s11596-019-2077-4 · Cited by 66 · cited 1×
“Studies have further applied AI for the classification of patients into diabetic or non-diabetic [15,16], the risk prediction of developing type 2 diabetes [17,18], diabetes diagnosis [19], and blood glucose level predictions [20].”
Westerdijk, K., Simons, K. S., Zegers, M., Wever, P. C., Pickkers, P., & de Jager, C. P. C. (2019) · PLoS ONE · doi:10.1371/journal.pone.0212861 · Cited by 65 · cited 2×
“…bacterial infection: >2 ng/mL [80] - In-hospital mortality predictor of COVID-19 patients: ≥ 0.56 ng/mL on ICU admission [73] WBC - Biomarker of infection [48] - Immature granulocytes [81] - NLR [82,83] - Non-infectious mimics, drugs, and comorbidities [48] Refer to Glossary section for acronyms.”
InfluentialApplication of mortality prediction systems to individual intensive care units
Patel, P., & Grant, B. J. B. (1999) · Intensive Care Medicine · doi:10.1007/s001340050992 · Cited by 64
Ho, K. M., DOBB, G. J., Lee, K. Y., Towler, S. C. B., & Webb, S. A. R. (2006) · Journal of Critical Care · doi:10.1016/j.jcrc.2006.01.005 · Cited by 64
Harerimana, G., Kim, J. W., & Jang, B. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2021.103778 · Cited by 64 · cited 2×
“%) studies [149,151,152,173,176,184], followed by MIMIC-IV [187] used by 4 (10”
Yadgarov, M. Y., Landoni, G., Бондоевич, Б. Л., Polyakov, P. A., Kadantseva, K., Smirnova, A. V., Kuznetsov, I. V., Shemetova, M. M., Yakovlev, A. А., & Likhvantsev, V. V. (2024) · Frontiers in Medicine · doi:10.3389/fmed.2024.1491358 · Cited by 64
Hu, C., Li, L., Li, Y., Wang, F., Hu, B., & Peng, Z. (2022) · Infectious Diseases and Therapy · doi:10.1007/s40121-022-00671-3 · Cited by 63 · cited 1×
“For sepsis mortality prediction models, built-in feature importance [113,240–242], SHAP [240,241,243–246], LIME [127,245,247], Break Down [241], Ceteris Paribus [241], and INVASE [242] are the explainable AI techniques used in literature.”
Host Diagnostic Biomarkers of Infection in the ICU: Where Are We and Where Are We Going?
Heffernan, A. J., & Denny, K. J. (2021) · Current Infectious Disease Reports · doi:10.1007/s11908-021-00747-0 · Cited by 63
Quintairos, A., Pilcher, D. V., & Salluh, J. I. (2022) · Intensive Care Medicine · doi:10.1007/s00134-022-06914-8 · Cited by 63
Early Detection of Septic Shock Onset Using Interpretable Machine Learners
Misra, D., Avula, V., Wolk, D. M., Farag, H. A., Li, J., Mehta, Y. B., Sandhu, R. K., Karunakaran, B., Kethireddy, S., Zand, R., et al. (2021) · Journal of Clinical Medicine · doi:10.3390/jcm10020301 · Cited by 62 · cited 5×
“Oversampling or down sampling data using different class balancing approaches (e.g., SMOTE [120,126,129,130,145], ADASYN [173]) on the training set, are a common preprocessing part, ensuring the model learns to detect both classes.”
InfluentialPredicting Prolonged Length of ICU Stay through Machine Learning
Wu, J., Lin, Y., Li, P., Hu, Y., Zhang, L., & Kong, G. (2021) · Diagnostics · doi:10.3390/diagnostics11122242 · Cited by 62 · cited 4×
“[171,175] use a DL model with AUC of 0.76 and 0.915, respectively, while 12 [151,159,168–170 , 172– 174”
InfluentialLi, Z., Wang, H., Liu, J., Chen, B., & Li, G. (2014) · Mediators of Inflammation · doi:10.1155/2014/641039 · Cited by 61 · cited 1×
“…[ C-Reactive Protein (CRP) is used as a biomarker for Systemic Inflammatory Response Syndrome (SIRS), infection and sepsis [49], for the diagnosis of neonatal sepsis [50], and for the differential diagnosis of bacterial versus viral infections [56,57] and their early identification [58,59].”
Coster, D., Wasserman, A., Fisher, E., Rogowski, O., Zeltser, D., Shapira, I., Bernstein, D. A., Meilik, A., Raykhshtat, E., Halpern, P., et al. (2019) · Infection · doi:10.1007/s15010-019-01383-6 · Cited by 60 · cited 2×
“However, it excludes cardiac surgery and myocardial infraction patients and MPM II also excludes patients below or at the age of 16 years old, and those with burn injuries and coronary artery disease [57,88,104].”
Persson, I., Östling, A., Arlbrandt, M., Söderberg, J., & Becedas, D. (2021) · JMIR Formative Research · doi:10.2196/28000 · Cited by 58 · cited 6×
“DL models proposed are usually temporal ones, like the Long Short Term Memory (LSTM) [122,127,135,143], the Recurrent Neural Network (RNN) [131,133], the Convolutional Neural Network (CNN) [115,140], the Temporal Convolutional Network (TCN) [118], or combinations of them [132,134].”
InfluentialBurdick, H., Pino, E., Gabel-Comeau, D., Gu, C., Roberts, J., Le, S., Slote, J., Saber, N., Pellegrini, E., Green‐Saxena, A., et al. (2020) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-020-01284-x · Cited by 58 · cited 4×
“Most of the ML models are tree-based ones like Extreme Gradient Boosting (XGB) [113,116,120,124,125,141], Random Forest (RF) [117,129,145], Gradient Boosting (GB) [114], Light Gradient Boosting Machine (LGBM) [119], and AdaBoost [144].”
InfluentialLee, J.-T., Hsieh, C.-C., Lin, C.-H., Lin, Y.-J., & Kao, C.-Y. (2021) · Scientific Reports · doi:10.1038/s41598-021-98961-2 · Cited by 58 · cited 1×
“[191,192,194,200,205] propose a DL models achieving AUC performance ranging from 0.80 to 0.9014.”
InfluentialDornbusch, H. J., Strenger, V., Sovinz, P., Lackner, H., Schwinger, W., Kerbl, R., & Urban, C. E. (2008) · Supportive Care in Cancer · doi:10.1007/s00520-007-0381-1 · Cited by 55 · cited 1×
“CRP is also associated with non-infection related aspects, like allergic complications [54], specific drug overdoses [54], obesity, smoking, diabetes mellitus, lack of exercise, hormonal therapy [64], and some hematological therapies [65].”
Bai, Y., Xia, J., Huang, X., Chen, S., & Zhan, Q. (2022) · Frontiers in Physiology · doi:10.3389/fphys.2022.1050849 · Cited by 54 · cited 1×
“Most of the ML models are tree-based ones like Extreme Gradient Boosting (XGB) [113,116,120,124,125,141], Random Forest (RF) [117,129,145], Gradient Boosting (GB) [114], Light Gradient Boosting Machine (LGBM) [119], and AdaBoost [144].”
Predicting length of stay in hospitals intensive care unit using general admission features
Abd-Elrazek, M. A., Eltahawi, A. A., Elaziz, M., & Abd-Elwhab, M. N. (2021) · Ain Shams Engineering Journal · doi:10.1016/j.asej.2021.02.018 · Cited by 53
Hong, Y., Wu, X., Qu, J., Gao, Y., Chen, H., & Zhang, Z. (2020) · Annals of Translational Medicine · doi:10.21037/atm.2020.03.147 · Cited by 52 · cited 1×
“Recurrent Feature Elimination (RFE) [123,173] and statistical tests [113,120,123,142,171,177,180,193,196] are commonly used for feature selection.”
Picod, A., Morisson, L., de Roquetaillade, C., Sadoune, M., Mebazaa, A., Gayat, É., Davison, B. A., Cotter, G., & Chousterman, B. G. (2022) · Frontiers in Immunology · doi:10.3389/fimmu.2022.868348 · Cited by 50 · cited 3×
“Interleukin-6 (IL-6) is used as an inflammation biomarker in septic and non-septic patients [74] and as a predictor of disease severity [75].”
Qu, R., Hu, L., Ling, Y., Hou, Y., Fang, H., Zhang, H., Liang, S., He, Z., Fang, M., Li, J., et al. (2020) · BMC Anesthesiology · doi:10.1186/s12871-020-01207-3 · Cited by 49 · cited 2×
“CRP has been proven to be a predictor of ICU mortality when more than 62.8 mg/L [55] and of severe COVID-19 in patients below 50 years old [49].”
InfluentialFenn, A., Davis, C., Buckland, D. M., Kapadia, N. N., Nichols, M., Gao, M. C., Knechtle, W. S., Balu, S., Sendak, M., & Theiling, J. (2021) · Annals of Emergency Medicine · doi:10.1016/j.annemergmed.2021.02.029 · Cited by 49 · cited 3×
“Other important features are previous visit outcomes [198], ambulance use [192], intravenous therapy (IVT) [193], and different lab values [203,204,208].”
Deep learning-based prediction of in-hospital mortality for sepsis
Yong, L., & Zhenzhou, L. (2024) · Scientific Reports · doi:10.1038/s41598-023-49890-9 · Cited by 48 · cited 3×
“The most important features according to built-in model importance are usually vital signs (e.g., heart rate, respiratory rate) [112,123,135,136,141] and laboratory values (e.g., Platelets, lactate) [112,113,117,119,124,126,141,142,145].”
Artificial Intelligence in the Intensive Care Unit: Current Evidence on an Inevitable Future Tool
Suresh, V., Singh, K. K., Vaish, E., Gurjar, M., Anubuvanan, A. M., Khulbe, Y., & Muzaffar, S. N. (2024) · Cureus · doi:10.7759/cureus.59797 · Cited by 48 · cited 1×
“While similar review studies exist, they focus on a single ICU outcome [36–43], explore studies that use a single type of data [36,38,42], do not elaborate on clinical approaches used in ICU [36–44], and/or do not significantly guide towards future work [36,37,43].”
Alabbad, D. A., Almuhaideb, A. M., Alsunaidi, S. J., Alqudaihi, K. S., Alamoudi, F. A., Alhobaishi, M. K., Alaqeel, N. A., & Alshahrani, M. S. S. (2022) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2022.100937 · Cited by 47 · cited 1×
“Most ML models are tree-based, and mainly RF [151,164,167,168,173, 174,176,178,180–182], XGB [170,172], and LGBM [150,163].”
Mowbray, F., Zargoush, M., Jones, A., de Wit, K., & Costa, A. P. (2020) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2020.104163 · Cited by 46 · cited 3×
“Other important features are previous visit outcomes [198], ambulance use [192], intravenous therapy (IVT) [193], and different lab values [203,204,208].”
Kim, T., Tae, Y., Yeo, H. J., Jang, J. H., Cho, K., Yoo, D., Lee, Y., Ahn, S., Kim, Y., Lee, N., et al. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12227156 · Cited by 44 · cited 7×
“[110–113, 115–122, 128– 130 In terms of features included, it is shown that kinematics features [122], free-text data [130,141] and a combination of hematological parameters [142] improve performance.”
InfluentialEvaluation Metrics for Generative Models: An Empirical Study
Betzalel, E., Penso, C., & Fetaya, E. (2024) · Machine Learning and Knowledge Extraction · doi:10.3390/make6030073 · Cited by 42 · cited 1×
“Additionally, there is strong interest in using Generative AI [251,252] to generate data for rare disease patients or for cases where text, image, and/or video data are limited.”
Sills, M. R., Ozkaynak, M., & Jang, H. (2021) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2021.104468 · Cited by 41 · cited 2×
“Other important features are previous visit outcomes [198], ambulance use [192], intravenous therapy (IVT) [193], and different lab values [203,204,208].”
Dan, T., Yang, L., Zhu, Z., Chen, X., Quan, W., Hu, Y., Tao, G., Zhu, L., Zhu, J., Jin, Y., et al. (2020) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm49941.2020.9313292 · Cited by 41
Cavallazzi, R., Bennin, C.-L., Hirani, A., Gilbert, C. R., & Marik, P. E. (2010) · Journal of Intensive Care Medicine · doi:10.1177/0885066610377980 · Cited by 39
Scoring Systems for the Patients of Intensive Care Unit
Jeong, S. (2018) · Acute and Critical Care · doi:10.4266/acc.2018.00185 · Cited by 39 · cited 1×
“SAPS II was validated only in North America and Europe and excluded patients below or at the age of 16 years old, as well as those with burn injuries, coronary artery disease, or a history of cardiac surgery [84,85,88,94,101–103].”
Jiang, Z., Bo, L., Wang, L., Xie, Y., Cao, J., Yao, Y., Lu, W., Deng, X., Yang, T., & Bian, J. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107772 · Cited by 38
A Deep Learning-Based Sepsis Estimation Scheme
Al‐Mualemi, B. Y., & Lu, L. (2020) · IEEE Access · doi:10.1109/access.2020.3043732 · Cited by 38 · cited 1×
“DL models proposed are usually temporal ones, like the Long Short Term Memory (LSTM) [122,127,135,143], the Recurrent Neural Network (RNN) [131,133], the Convolutional Neural Network (CNN) [115,140], the Temporal Convolutional Network (TCN) [118], or combinations of them [132,134].”
Yahya, B. M., Yahya, F. S., & Thannoun, R. G. (2021) · Applied Geomatics · doi:10.1007/s12518-021-00365-4 · Cited by 38 · cited 1×
“ML techniques have been used to predict whether patients are infected by COVID-19 [30], identify its spread [31,32], predict the number of discharged patients and deaths [33], and predict COVID-19 mortality [34,35] to prioritize triage and hospitalization [35].”
An interpretable risk prediction model for healthcare with pattern attention
Kamal, S. A., Yin, C., Qian, B., & Zhang, P. (2020) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-020-01331-7 · Cited by 36 · cited 1×
“[112,115,118,123,126–128, 130– 137, 139, In summary, we differentiate between models used for detecting sepsis presence versus the prediction of sepsis at a future time.”
Ahmed, A., Ashour, O. M., Ali, H., & Firouz, M. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117314 · Cited by 36
Sepsis Mortality Prediction Using Wearable Monitoring in Low–Middle Income Countries
Ghiasi, S., Zhu, T., Lü, P., Hagenah, J., Khanh, P. N. Q., Hảo, N. V., Thwaites, L., & Clifton, D. A. (2022) · Sensors · doi:10.3390/s22103866 · Cited by 36
Gülcher, S. S., Bruins, N. A., Kingma, W. P., & Boerma, E. C. (2016) · Annals of Intensive Care · doi:10.1186/s13613-016-0105-0 · Cited by 35
Chen, Q., Zhang, B., Yang, J., Mo, X., Zhang, L., Li, M., Chen, Z., Jin, F., Wang, F., Huang, W., et al. (2021) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2021.675431 · Cited by 35 · cited 3×
“Papers found are split in 3 LOS outcome categories: (i) a continuous outcome in hours/days [149–167] (19, 49%), (ii) a binary outcome of LOS > X days [151,159,168–179] (14, 36%), and (iii) a multiclass outcome of LOS > X days [152,163,180–185] (8, 21%).”
Lin, P.-C., Chen, K.-T., Chen, H.-C., Islam, M. M., & Lin, M. (2021) · Journal of Personalized Medicine · doi:10.3390/jpm11111055 · Cited by 35 · cited 4×
“[116,124,141,142,145] of them used a ML model with AUC ranging from 0.86 to 0.97, while just 1 study [122] uses a DL model with an AUC of 0.835.”
InfluentialPyHealth: A Python Library for Health Predictive ModelsPreprint
Zhao, Y., Qiao, Z., Xiao, C., Glass, L. M., & Sun, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.04209 · Cited by 35 · cited 2×
“…downloadable pipelines that predict LOS, among other clinical outcomes, using a range of models and databases are given in [111] https://github.com/rvandewater/YAIB (accessed on 19 July 2024) and [146] https://github.com/yzhao062/PyHealth (accessed on 19 July 2024) which can handle multimodal data.”
Early prediction of sepsis using double fusion of deep features and handcrafted features
Duan, Y., Huo, J., Chen, M., Hou, F., Yan, G., Li, S., & Wang, H. (2023) · Applied Intelligence · doi:10.1007/s10489-022-04425-z · Cited by 34 · cited 1×
“[112,115,118,123,126–128, 130– 137, 139, In summary, we differentiate between models used for detecting sepsis presence versus the prediction of sepsis at a future time.”
A robust and parsimonious machine learning method to predict ICU admission of COVID-19 patients
Famiglini, L., Campagner, A., Carobene, A., & Cabitza, F. (2022) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-022-02543-x · Cited by 33 · cited 1×
“Other important features are previous visit outcomes [198], ambulance use [192], intravenous therapy (IVT) [193], and different lab values [203,204,208].”
Gholamzadeh, M., Abtahi, H., & Safdari, R. (2023) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2023.101236 · Cited by 32 · cited 4×
“[112–129] use the Sepsis-3 definition [98].”
InfluentialBrann, F., Sterling, N. W., Frisch, S. O., & Schrager, J. D. (2023) · JMIR AI · doi:10.2196/49784 · Cited by 30 · cited 5×
“Most of the ML models are tree-based ones like Extreme Gradient Boosting (XGB) [113,116,120,124,125,141], Random Forest (RF) [117,129,145], Gradient Boosting (GB) [114], Light Gradient Boosting Machine (LGBM) [119], and AdaBoost [144].”
InfluentialArtificial Intelligence-Based Models for Prediction of Mortality in ICU Patients: A Scoping Review
Olang, O., Mohseni, S., Shahabinezhad, A., Hamidianshirazi, Y., Goli, A. N., Abolghasemian, M., Shafiee, M. A., Aarabi, M., Alavinia, M., & Shaker, P. (2024) · Journal of Intensive Care Medicine · doi:10.1177/08850666241277134 · Cited by 30 · cited 1×
“While similar review studies exist, they focus on a single ICU outcome [36–43], explore studies that use a single type of data [36,38,42], do not elaborate on clinical approaches used in ICU [36–44], and/or do not significantly guide towards future work [36,37,43].”
Barsasella, D., Gupta, S., Malwade, S., Aminin, Susanti, Y., Tirmadi, B., Mutamakin, A., Jonnagaddala, J., & Syed-Abdul, S. (2021) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2021.104569 · Cited by 29 · cited 2×
“Papers found are split in 3 LOS outcome categories: (i) a continuous outcome in hours/days [149–167] (19, 49%), (ii) a binary outcome of LOS > X days [151,159,168–179] (14, 36%), and (iii) a multiclass outcome of LOS > X days [152,163,180–185] (8, 21%).”
Zhong, H., Wang, B., Wang, D., Liu, Z., Xing, C., Wu, Y. C., Gao, Q., Zhu, S., Qu, H., Jia, Z., et al. (2021) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2021.104572 · Cited by 29
Transferability and interpretability of the sepsis prediction models in the intensive care unit
Chen, Q., Li, R., Lin, C., Lai, C., Chen, D., Qu, H., Huang, Y., Lu, W., Tang, Y., & Li, L. (2022) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-022-02090-3 · Cited by 28 · cited 1×
“For sepsis prediction, most studies use ML built-in feature importance [112,117,119,124,129,135,136,141,232,233] and SHAP [119,128,233–236].”
Goodlet, K. J., Cameron, E. A., & Nailor, M. D. (2020) · Open Forum Infectious Diseases · doi:10.1093/ofid/ofaa096 · Cited by 28 · cited 1×
“However, it excludes cardiac surgery and myocardial infraction patients and MPM II also excludes patients below or at the age of 16 years old, and those with burn injuries and coronary artery disease [57,88,104].”
Multi-modal learning for inpatient length of stay prediction
Chen, J., Wen, Y., Pokojovy, M., Tseng, T.-L., McCaffrey, P., Vo, A. H., Walser, E., & Moen, S. T. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108121 · Cited by 28 · cited 2×
“%) studies [161,162,175,178], and eICU [188] used by 4 (10”
An Extensive Data Processing Pipeline for MIMIC-IV
Gupta, M., Gallamoza, B., Cutrona, N., Dhakal, P., Poulain, R., & Beheshti, R. (2022) · arXiv · doi:10.48550/arxiv.2204.13841 · Cited by 28 · cited 3×
“[171,175] use a DL model with AUC of 0.76 and 0.915, respectively, while 12 [151,159,168–170 , 172– 174”
InfluentialAlabdulhafith, M., Saleh, H., Elmannai, H., Ali, Z. H., El-Sappagh, S., Hu, J.-W., & El-Rashidy, N. (2023) · IEEE Access · doi:10.1109/access.2023.3312343 · Cited by 27
Seo, H., Ahn, I., Gwon, H., Kang, H. J., Kim, Y., Cho, H. N., Choi, H., Kim, M., Han, J., Kee, G., et al. (2023) · Health Care Management Science · doi:10.1007/s10729-023-09660-5 · Cited by 27 · cited 2×
“More commonly used Convolutional Neural Networks, either for time series, image, or video data, are now substituted/complemented by Transformer-based models [202].”
Abujaber, A. A., Fadlalla, A. M., Nashwan, A. J., El-Menyar, A. A., & Al-Thani, H. A. S. (2022) · Intelligence-Based Medicine · doi:10.1016/j.ibmed.2022.100052 · Cited by 26 · cited 3×
“[171,175] use a DL model with AUC of 0.76 and 0.915, respectively, while 12 [151,159,168–170 , 172– 174”
InfluentialLi, S., Dou, R., Song, X., Lui, K., Xu, J., Guo, Z., Hu, X., Guan, X., & Cai, C. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12030915 · Cited by 25 · cited 1×
“For sepsis mortality prediction models, built-in feature importance [113,240–242], SHAP [240,241,243–246], LIME [127,245,247], Break Down [241], Ceteris Paribus [241], and INVASE [242] are the explainable AI techniques used in literature.”
Narin, A., İşler, Y., & Özer, M. (2016) · Medical Technologies Congress (TIPTEKNO) · doi:10.1109/tiptekno.2016.7863110 · Cited by 25 · cited 1×
“Regarding cardiovascular diseases, AI has been used to assess the risk of developing a cardiovascular disease [21–27] and identify heart rate severity [28,29].”
Lin, Y.-T., Deng, Y.-X., Tsai, C., Huang, C., & Fu, L. (2024) · JMIR Medical Informatics · doi:10.2196/48862 · Cited by 25 · cited 3×
“Text data are often included [200,202,205], with a study suggesting that the use of text can improve performance [202].”
InfluentialLi, Q., & Gong, X. (2018) · Experimental and Therapeutic Medicine · doi:10.3892/etm.2018.5960 · Cited by 25 · cited 5×
“It is associated with both infection and inflammation [48], viral and bacterial infections, and sepsis [79], as well as with the severity of infection and risk of death [78].”
InfluentialA prospective evaluation of the Infection Probability Score (IPS) in the intensive care unit
Martini, A., Gottin, L., Mélot, C., & Vincent, J. (2008) · Journal of Infection · doi:10.1016/j.jinf.2008.02.015 · Cited by 24 · cited 2×
“It is a component of the International Patient Summary (IPS) [51,52] and of ICU prognostic blood tests [53].”
Towards an Explainable Model for Sepsis Detection Based on Sensitivity Analysis
Chen, M., & Hernández, A. (2022) · IRBM · doi:10.1016/j.irbm.2021.05.006 · Cited by 23 · cited 2×
“Sensitivity analysis [232,237], LIME [127,238], heatmaps [239] and Grad-CAM [234] are also used.”
Deliberate clinical inertia: Using meta‐cognition to improve decision‐making
Keijzers, G., Fatovich, D. M., Egerton‐Warburton, D., Cullen, L., Scott, I. A., Glasziou, P., & Croskerry, P. (2018) · Emergency Medicine Australasia · doi:10.1111/1742-6723.13126 · Cited by 23
Doğu, E., Albayrak, Y. E., & Tuncay, E. (2021) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-021-02327-9 · Cited by 23
Bhadouria, A. S., & Singh, R. K. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16474-8 · Cited by 23 · cited 1×
“Papers found are split in 3 LOS outcome categories: (i) a continuous outcome in hours/days [149–167] (19, 49%), (ii) a binary outcome of LOS > X days [151,159,168–179] (14, 36%), and (iii) a multiclass outcome of LOS > X days [152,163,180–185] (8, 21%).”
Chang, J., Canêo, L. F., Turquetto, A. L. R., Amato, L. P., Arita, E. C. T. C., Fernandes, A. M. D. S., Trindade, E. M., Jatene, F. B., Dossou, P.-E., & Jatene, M. B. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e25406 · Cited by 21 · cited 2×
“Papers found are split in 3 LOS outcome categories: (i) a continuous outcome in hours/days [149–167] (19, 49%), (ii) a binary outcome of LOS > X days [151,159,168–179] (14, 36%), and (iii) a multiclass outcome of LOS > X days [152,163,180–185] (8, 21%).”
Choi, J.-S., Trinh, T. X., Ha, J., Yang, M.-S., Lee, Y., Kim, Y.-E., Choi, J., Byun, H., Song, J., & Yoon, T. H. (2020) · Scientific Reports · doi:10.1038/s41598-019-57107-1 · Cited by 20
Chen, T., Huang, T.-Y., & Chang, Y. (2024) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2024.104657 · Cited by 20 · cited 3×
“Text data are often included [200,202,205], with a study suggesting that the use of text can improve performance [202].”
InfluentialCOP-E-CAT: cleaning and organization pipeline for EHR computational and analytic tasks
Mandyam, A., Yoo, E. C., Soules, J., Laudański, K., & Engelhardt, B. E. (2021) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3459930.3469536 · Cited by 19
Research works printed but not linked: 12
- Isravel, 2020, Improved heart disease diagnostic IoT model using machine learning techniques, Neuroscience, 9, 4442
- Gupta, 2021, AI-Enabled COVID-19 Outbreak Analysis and Prediction: Indian States Vs. Union Territories, Comput. Mater. Contin., 67, 933
- Flores, 2001, C reactive protein as marker of infection among patients with severe closed trauma, Enfermedades Infecc. Microbiol. Clin., 19, 61
- Ayazoglu, 2011, A comparison of APACHE II and APACHE IV scoring systems in predicting outcome in patients admitted with stroke to an intensive care unit, Anaesth. Pain Intensiv. Care, 15, 7
- Yang, J., Karstens, L., Ross, C., and Yala, A. (2022). AI gone astray: Technical supplement. arXiv.
- Rocheteau, E., Tong, C., Veličković, P., Lane, N., and Liò, P. (2021). Predicting patient outcomes with graph representation learning. arXiv.
- Wen, 2022, Time-to-event modeling for hospital length of stay prediction for COVID-19 patients, Mach. Learn. Appl., 9, 100365
- Ahmed, 2024, Developing a decision model to early predict ICU admission for COVID-19 patients: A machine learning approach, Intell. Med., 9, 100136
- Lundberg, S., and Lee, S.-I. (2017). SHAP: A Unified Approach to Interpreting Model Predictions. Adv. Neural Inf. Process. Syst., 1–10.
- Pick, F. (2021). Explainable Machine Learning for Predicting Sepsis Outcome. [Master’s Thesis, Department Computer Science at Swansea University].
- Singhal, K., Tu, T., Gottweis, J., Sayres, R., Wulczyn, E., Hou, L., Clark, K., Pfohl, S., Cole-Lewis, H., and Neal, D. (2023). Towards expert-level medical question answering with large language models. arXiv.
- Team, G., Georgiev, P., Lei, V.I., Burnell, R., Bai, L., Gulati, A., Tanzer, G., Vincent, D., Pan, Z., and Wang, S. (2024). Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv.
Cited by
Abbas, Q., Jeong, W., & Lee, S. W. (2025) · Healthcare · doi:10.3390/healthcare13172154 · Cited by 236
“This not only aids in clinical interpretation but also aligns AI recommendations with clinical reasoning, thereby increasing user trust and adoption [24,25].”
Transforming sepsis management: AI-driven innovations in early detection and tailored therapies
Papareddy, P., Lobo, T. J., Holub, M., Bouma, H., Maca, J., Strodthoff, N., & Herwald, H. (2025) · Critical Care · doi:10.1186/s13054-025-05588-0 · Cited by 49
“While several reviews have explored the promise of AI in sepsis management, focus mainly either on early detection models [21, 22] or on specific algorithmic techniques [23, 24], but to a less extend aspects such as clinical translation, regulatory evaluation, and real-world implementation.”
Artificial intelligence and infectious diseases: Scope and perspectives
Abbara, S., Crabol, Y., de Bouillé, J. G., Dinh, A., & Morquin, D. (2025) · Infectious Diseases Now · doi:10.1016/j.idnow.2025.105131 · Cited by 20
Recent approaches of artificial intelligence in intensive care unit: A review
Havelikar, U. A., Bembde, A., Patil, D., Shinde, A., Sharma, R. K., Patel, V. P., & Chaudhari, N. B. (2025) · Intelligent Hospital · doi:10.1016/j.inhs.2025.100030 · Cited by 17
Athukorala, V. S., & Ilmini, W. M. K. S. (2026) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-026-03344-0 · Cited by 10
“Furthermore, Stylianides et al. [38] present significant advancements in AI applications within intensive care units, emphasizing sepsis prediction.”
Metadata
- Authors
- Stylianides, C., Nicolaou, A., Sulaiman, W. A., Alexandropoulou, C.-A., Panagiotopoulos, I., Karathanasopoulou, K., Dimitrakopoulos, G., Kleanthous, S., Politi, E., Ntalaperas, D., Papageorgiou, X., Garcia, F., Antoniou, Z., Ioannides, N., Palazis, L., Vavlitou, A., Pattichis, M. S., Pattichis, C. S., & Panayides, A. S.
- Affiliations
- Centre of Excellence, CYENS, Nicosia 1016, Cyprus
- Harokopio University of Athens (GR)
- Department of Informatics and Telematics, Harokopio University of Athens, 176 76 Kallithea, Greece
- Open University of Cyprus (CY)
- UBITECH Limited, Limassol 3071, Cyprus
- Research & Development Department, 3aHealth, Strovolos 2020, Cyprus
- State Health Services Organization, Aglantzia 2100, Cyprus
- University of New Mexico (US)
- Corresponding author
- Stylianides, Charithea
- ORCID
- Venue
- Machine Learning and Knowledge Extraction
- Publisher
- MDPI AG
- Published
- 2025-01-08
- Volume
- 7
- Issue
- 1
- Pages
- 6
- ISSN
- 2504-4990 (ISSN-L)
- OpenAlex
- W4406170188
- Semantic Scholar
- fcfd620a9a6093506eb00d2d7771e830231b33ae
- Cited by
- 28
- References
- 237
- Citations by year
- 2026: 16 · 2025: 12
- FWCI
- 19.52
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Machine Learning and Knowledge Extraction (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Republic of Cyprus (CODEVELOP-ICTHEALTH/0322/0071)
- Topics
- Health Sciences › Medicine › Critical Care and Intensive Care Medicine › Sepsis Diagnosis and Treatment
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- sepsis prediction, intensive care unit, artificial intelligence in healthcare, sepsis mortality prediction, hospital length of stay prediction, ICU admission prediction, emergency department admission, diagnostic biomarkers, explainable AI, healthcare resource allocation, multimodal data, transformer models
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.