Optimizing Autoencoders for Learning Deep Representations from Health Data
Abstract
Analyzing patients' health data using machine learning techniques can improve both patient outcomes and hospital operations. However, heterogeneous patient data (e.g., vital signs) and inefficient feature learning methods affect the implementation of machine learning-based patient data analysis. In this paper, we present a novel unsupervised deep learning-based feature learning (DFL) framework to automatically learn compact representations from patient health data for efficient clinical decision making. Real-world pneumonia patient data from the National University Hospital in Singapore are collected and analyzed to evaluate the performance of DFL. Furthermore, publicly available electroencephalogram data are extracted from the UCI Machine Learning Repository to test and support our findings. Using both data sets, we compare the performance of DFL to that of several popular feature learning methods and demonstrate its advantages.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
For outcomes that happen in hospital, such as inpatient mortality, length-of-stay, and intervention prediction, observation windows at hour level are developed.
“On the contrary, for those outcomes that happen in hospital, such as inpatient mortality [14,57,61,65], length-of-stay [14,83], and intervention prediction [32,55,65], observation windows at hour level are developed.”
From Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review (2021) · cites this paper as [83]
Cites
LIBSVM: A library for support vector machines
Chang, C.-C., & Lin, C.-J. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1961189.1961199 · Cited by 40,918 · see citing works
Representation Learning: A Review and New Perspectives
Bengio, Y., Courville, A., & Vincent, P. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.50 · Cited by 13,001 · see citing works
Gaussian Processes for Machine Learning
Rasmussen, C. E., & Williams, C. K. I. (2005) · doi:10.7551/mitpress/3206.001.0001 · Cited by 11,705 · see citing works
Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation
Powers, D. M. W. (2020) · arXiv · doi:10.48550/arxiv.2010.16061 · Cited by 5,440 · see citing works
(2010) · Journal of Machine Learning Research · doi:10.5555/1756006.1953039 · Cited by 4,051 · see citing works
Miotto, R., Li, L., Kidd, B. A., & Dudley, J. T. (2016) · Scientific Reports · doi:10.1038/srep26094 · Cited by 1,710 · see citing works
Mining electronic health records: towards better research applications and clinical care
Jensen, P. B., Jensen, L. J., & Brunak, S. (2012) · Nature Reviews Genetics · doi:10.1038/nrg3208 · Cited by 1,680 · see citing works
Neural networks and principal component analysis: Learning from examples without local minima
Baldi, P., & Hornik, K. (1989) · Neural Networks · doi:10.1016/0893-6080(89)90014-2 · Cited by 1,383 · see citing works
Weiskopf, N. G., & Weng, C. (2013) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2011-000681 · Cited by 1,216 · cited 1× · see citing works
“ments and the clinical tests conducted will vary among different patients [6], [15].”
Chalfin, D. B., Trzeciak, S., Likourezos, A., Baumann, B. M., & Dellinger, R. P. (2007) · Critical Care Medicine · doi:10.1097/01.ccm.0000266585.74905.5a · Cited by 1,022 · see citing works
Compressed Sensing for Real-Time Energy-Efficient ECG Compression on Wireless Body Sensor Nodes
Mamaghanian, H., Khaled, N., Atienza, D., & Vandergheynst, P. (2011) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2011.2156795 · Cited by 701 · see citing works
Beyond Atopy: Multiple Patterns of Sensitization in Relation to Asthma in a Birth Cohort Study
Simpson, A., Tan, V. Y. F., Winn, J., Svensén, M., Bishop, C. M., Heckerman, D. E., Buchan, I., & Custovic, A. (2010) · American Journal of Respiratory and Critical Care Medicine · doi:10.1164/rccm.200907-1101oc · Cited by 415 · see citing works
Unsupervised learning of disease progression models
Wang, X., Sontag, D. A., & Wang, F. (2014) · ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/2623330.2623754 · Cited by 220 · see citing works
MISSING DATA IMPUTATION IN THE ELECTRONIC HEALTH RECORD USING DEEPLY LEARNED AUTOENCODERS
Beaulieu‐Jones, B. K., & Moore, J. H. (2016) · Pacific Symposium on Biocomputing (PSB) · doi:10.1142/9789813207813_0021 · Cited by 194 · see citing works
Deep learning for healthcare decision making with EMRs
Liang, Z., Zhang, G., Huang, J. X., & Hu, Q. V. (2014) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm.2014.6999219 · Cited by 187 · see citing works
Using artificial intelligence to improve hospital inpatient care
Neill, D. B. (2013) · IEEE Intelligent Systems · doi:10.1109/mis.2013.51 · Cited by 177 · see citing works
Semi-supervised learning of the electronic health record for phenotype stratification
Beaulieu‐Jones, B. K., & Greene, C. S. (2016) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2016.10.007 · Cited by 175 · see citing works
Learning and generalization with the information bottleneck
Shamir, O., Sabato, S., & Tishby, N. (2010) · Theoretical Computer Science · doi:10.1016/j.tcs.2010.04.006 · Cited by 171 · see citing works
Rincón, F., Mayer, S. A., Rivolta, J., Stillman, J. I., Boden-Albala, B. M., Elkind, M. S. V., Marshall, R. S., & Chong, J. Y. (2010) · Neurocritical Care · doi:10.1007/s12028-010-9347-0 · Cited by 139 · see citing works
Nutritional parameters associated with prolonged hospital stay among ambulatory adult patients
Caccialanza, R., Klersy, C., Cereda, E., Cameletti, B., Bonoldi, A., Bonardi, C., Marinelli, M., & Dionigi, P. (2010) · Canadian Medical Association Journal · doi:10.1503/cmaj.091977 · Cited by 113 · see citing works
Toward personalizing treatment for depression: predicting diagnosis and severity
Huang, S. H., LePendu, P., Iyer, S. V., Tai‐Seale, M., Carrell, D., & Shah, N. H. (2014) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2014-002733 · Cited by 108 · see citing works
Identifying and mitigating biases in EHR laboratory tests
Pivovarov, R., Albers, D. J., Sepulveda, J. L., & Elhadad, N. (2014) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2014.03.016 · Cited by 91 · see citing works
A Framework for Mining Signatures from Event Sequences and Its Applications in Healthcare Data
Wang, F., Lee, N., Hu, J., Sun, J., Ebadollahi, S., & Laine, A. F. (2012) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2012.111 · Cited by 89 · see citing works
Analysis of alcoholic EEG signals based on horizontal visibility graph entropy
Zhu, G., Li, Y., Wen, P. P., & Wang, S. (2014) · Brain Informatics · doi:10.1007/s40708-014-0003-x · Cited by 80 · see citing works
An integrated data mining approach to real-time clinical monitoring and deterioration warning
Mao, Y., Chen, W., Chen, Y., Lu, C., Kollef, M., & Bailey, T. (2012) · Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/2339530.2339709 · Cited by 76 · see citing works
Identifying phenotypic signatures of neuropsychiatric disorders from electronic medical records
Lyalina, S., Percha, B., LePendu, P., Iyer, S. V., Altman, R. B., & Shah, N. H. (2013) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2013-001933 · Cited by 61 · see citing works
Redesigning Intensive Care Unit Flow Using Variability Management to Improve Access and Safety
Ryckman, F. C., Yelton, P. A., Anneken, A. M., Kiessling, P., Schoettker, P. J., & Kotagal, U. R. (2009) · The Joint Commission Journal on Quality and Patient Safety · doi:10.1016/s1553-7250(09)35073-4 · Cited by 45 · see citing works
Spatio-temporal autoencoder for feature learning in patient data with missing observations
Jia, Y., Zhou, C., & Motani, M. (2017) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm.2017.8217773 · Cited by 25 · see citing works
Splines and variational methods, M. Prenter, Wiley, New York, 1975. £10.75
Zienkiewicz, O. C. (1976) · International Journal for Numerical Methods in Engineering · doi:10.1002/nme.1620100222 · Cited by 19 · see citing works
Learning Deep Representations from Heterogeneous Patient Data for Predictive Diagnosis
Zhou, C., Jia, Y., Motani, M., & Chew, J. (2017) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3107411.3107433 · Cited by 12 · see citing works
Data selection in EEG signals classification
Wang, S., Li, Y., Wen, P. P., & Lai, D. (2016) · Australasian Physical & Engineering Sciences in Medicine · doi:10.1007/s13246-015-0414-x · Cited by 10 · see citing works
Flagged first, then by citations
3 of the 34 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Si, Y., Du, J., Li, Z., Jiang, X., Miller, T., Wang, F., Jim Zheng, W., & Roberts, K. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2020.103671 · Cited by 235 · see citing works
Deep Learning Identifies Digital Biomarkers for Self-Reported Parkinson's Disease
Zhang, H., Deng, K., Li, H., Albin, R. L., & Guan, Y. (2020) · Patterns · doi:10.1016/j.patter.2020.100042 · Cited by 74 · see citing works
Industrial IoT Intrusion Detection via Evolutionary Cost-Sensitive Learning and Fog Computing
Telikani, A., Shen, J., Yang, J., & Wang, P. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3188224 · Cited by 68 · see citing works
“AEs have the ability to learn compact representations in an unsupervised learning schema [28].”
BackgroundWang, M., Wei, Z., Jia, M., Chen, L., & Ji, H. (2022) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-022-01776-y · Cited by 60 · see citing works
“Owing to high-dimensional sparse data (that is, data with more 0 values), the computational complexity in training is relatively high and the model is difficult to optimize [26].”
BackgroundKashef, A., Khatibi, T., & Mehrvar, A. (2020) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2020.100399 · Cited by 34 · see citing works
A Cost-Sensitive Machine Learning Model With Multitask Learning for Intrusion Detection in IoT
Telikani, A., Esmi, N., Soleymanpour, S., Shahbahrami, A., Shen, J., Gaydadjiev, G. N., & Hassanpour, R. (2023) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2023.3314208 · Cited by 27 · see citing works
“In an unsupervised learning setting, AEs are capable of learning compact representations [28].”
BackgroundWen, G., Chen, H., Li, H., Hu, Y., Li, Y., & Wang, C. (2020) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2020.103608 · Cited by 27 · see citing works
Patient Health Representation Learning via Correlational Sparse Prior of Medical Features
Ma, X., Wang, Y., Chu, X., Ma, L., Tang, W., Zhao, J., Yuan, Y., & Wang, G. (2022) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2022.3230454 · Cited by 26 · see citing works
“ing into real-world clinical tasks, including mortality predictions [1], [2], [3], [7], [14], [24], [25], diagnosis predictions [4], [5], [6], [7], [17], [24], [26], length-of-stay predictions [7], [8], [24], [27], patient subtyping [9], [28], and automatic”
An Alzheimer’s disease category progression sub-grouping analysis using manifold learning on ADNI
van der Haar, D. T., Moustafa, A. A., Warren, S. L., Alashwal, H., & van Zyl, T. L. (2023) · Scientific Reports · doi:10.1038/s41598-023-37569-0 · Cited by 23 · see citing works
“In turn, Zhou et al. 55 found that their autoencoder models could predict hospitalisation length with an accuracy of 71.4% and classify alcoholism with an accuracy of 65.1%.”
BackgroundMethodInfluentialYun, W. J., Shin, M., Jung, S., Ko, J., Lee, H., & Kim, J. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.106739 · Cited by 21 · see citing works
DeepHealth: Review and challenges of artificial intelligence in health informaticsPreprint
Kwak, G. H., & Hui, P. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.00384 · Cited by 38 · see citing works
“thods were used to either get dimensionality reduction or latent representation for the patient especially with words such as ICD-9, CPT, LOINC, NDC, procedure codes and diagnostic codes. Zhou et al. [309] analyzed patients’ health data, using unsupervised deep learning-based feature learning (DFL) framework to automa…”
MethodAssociation Rule-Based Breast Cancer Prevention and Control System
Li, A., Liu, L., Ullah, A., Wang, R., Ma, J., Huang, R., Yu, Z., & Ning, H. (2019) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2019.2912629 · Cited by 14 · see citing works
“In addition to medical image classification, some health research questions also use machine-learning methods [21]–[23].”
BackgroundŞentürk, Ü. (2025) · Ain Shams Engineering Journal · doi:10.1016/j.asej.2025.103436 · Cited by 10 · see citing works
Oh, B., Hwang, J., Seo, S., Chun, S., & Lee, K.-H. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.114912 · Cited by 8 · see citing works
Adaptation of Autoencoder for Sparsity Reduction From Clinical Notes Representation Learning
Le, T.-D., Noumeir, R., Rambaud, J., Sans, G., & Jouvet, P. A. (2023) · IEEE Journal of Translational Engineering in Health and Medicine · doi:10.1109/jtehm.2023.3241635 · Cited by 8 · see citing works
“For example, a study [13] demonstrates an efficient framework for automatically learning compact representations from heterogeneous raw data sources from patient health data.”
BackgroundJoint Triplet Autoencoder for histopathological colon cancer nuclei retrieval
Singh, S. R., Dubey, S. R., Ms, S., Ventrapragada, S., & Dasharatha, S. S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15400-2 · Cited by 7 · see citing works
“The autoencoder has been also used in medical area such as Convolutional Autoencoders to study the Alzheimer’s Disease [33] and Optimizing autoencoders based network to analyze health data [57].”
Deep Learning Radar for High-Fidelity Heart Sound Recovery in Real-World Scenarios
Tang, H., Rong, Y., Chai, L., & Bliss, D. W. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3288764 · Cited by 6 · see citing works
Varalakshmi, P., Yuvaraj, G., Dhanasekaran, K., & Samannan, K. (2018) · International Conference on Advanced Computing · doi:10.1109/icoac44903.2018.8939073 · Cited by 5 · see citing works
“learning-based Feature Learning framework to automatically learn compact representations from patient health data for efficient clinical decision making was proposed in [13] which used real world pneumonia patient data for evaluating the performance of the framework.”
MethodImproving Mental Health Support in Engineering Education using Machine Learning and Eye-Tracking
Liu, Y., Lu, W., Zavareh, A. T., Rigsby, M., & Zoghi, B. (2023) · IEEE Frontiers in Education Conference (FIE) · doi:10.1109/fie58773.2023.10343428 · Cited by 3 · see citing works
“However, the multiple hidden layers of nodes can make it challenging to interpret how the algorithm arrived at its output [24-28].”
BackgroundComputational Intelligence and Deep Learning in Health Informatics
Naskath, J., Rajakumari, R. G., Aldabbas, H., & Mustafa, Z. (2024) · doi:10.1002/9781394214259.ch9 · Cited by 3 · see citing works
A Novel Deep Learning Representation for Industrial Control System Data
Zhang, B., Shi, Y., Zhao, J., Wang, T., & Wang, K. (2023) · Intelligent Automation & Soft Computing · doi:10.32604/iasc.2023.033762 · Cited by 1 · see citing works
Multichannel Signal Classification Using Vector Autoregression
Haboub, A., Baali, H., & Bouzerdoum, A. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9054144 · Cited by 1 · see citing works
“In [21], a deep learning approach consisting of a stacked denoising autoencoder (SDAE) is employed.”
MethodPrediction of Breast Cancer Using Electronic Health Record
Poonguzhali, N., Dharani, V., Nivedha, R., & Ruby, L. S. (2020) · International Conference on System, Computation, Automation and Networking (ICSCAN) · doi:10.1109/icscan49426.2020.9262398 · Cited by 1 · see citing works
“[23] C.”
Gikera, R., Kenya, N., Mambo, S. M., Kenya, N., Mwaura, J., & Kenya, N. (2020) · International Conference on Intelligent and Innovative Computing Applications · doi:10.1145/3415088.3415125
2019 Index IEEE Journal of Biomedical and Health Informatics Vol. 23
(2019) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.2966416
Self-supervised deep learning and EEG categorization
Svantesson, M. (2024) · Linköping University medical dissertations · doi:10.3384/9789180755429
The Synergy of Autoencoders and Electronic Health Records for Improved Patient Care
Alkhawaji, R. N. (2025) · International Conference on Computing and Information Technology (ICCIT) · doi:10.1109/iccit63348.2025.10989336
“In the context of healthcare, EHRs contain vast amounts of data regarding a patient's health [25].”
Shen, Y., Zhang, T., Liu, Z., Kostelidou, K., Xu, Y., & Yang, L. (2025) · Communications Biology · doi:10.1038/s42003-025-08817-y
“Their ability to learn from high-dimensional data makes them well suited for analyzing the intricate dynamics involved in phenotypic expression 19,20 .”
Das, T., Ghosh, S., & Mishra, M. (2026) · International Conference on Recent Advancement in Electrical, Computer and Communication Technologies (IECCT) · doi:10.1109/iecct68664.2026.11541770
“Chen et al. [3] proposed an autoencoder-based framework for anomaly detection using reconstruction error, while Zhou et al. [4] improved feature extraction using optimized autoencoder architectures.”
- Authors
- Zhou, C., Jia, Y., & Motani, M.
- Affiliations
- National University of Singapore (SG)
- Venue
- IEEE Journal of Biomedical and Health Informatics
- Published
- 2019-01
- Volume
- 23
- Issue
- 1
- Pages
- 103-111
- ISSN
- 2168-2194 (ISSN-L)
- 2168-2208
- OpenAlex
- W2883672015
- Cited by
- 29
- References
- 34
- Citations by year
- 2026: 1 · 2025: 3 · 2024: 2 · 2023: 8 · 2022: 3 · 2021: 1 · 2020: 7 · 2019: 3 · 2018: 1
- FWCI
- 1.69
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Journal of Biomedical and Health Informatics (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Ministry of Education - Singapore (R-263-000-C81-114)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Keywords
- autoencoder, deep feature learning, unsupervised feature learning, patient health data, clinical decision support, pneumonia, EEG, vital signs, data heterogeneity, compact representation
- MeSH terms
- Machine Learning, Algorithms, Electroencephalography, Humans, Medical Informatics, Signal Processing, Computer-Assisted, Electronic Health Records
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
