CoChat
Sign Up
Cited by 1,636Artificial Intelligence in Healthcare and EducationArtificial Intelligence in HealthcareInternet of Things and AI

The rise of artificial intelligence in healthcare applications

Open access
Bohr, Adam, Memarzadeh, Kaveh·2020·Artificial Intelligence in Healthcare
DOI 10.1016/b978-0-12-818438-7.00002-2PMC PMC7325854
Open this paper in CoChat — free account

Abstract

Big data and machine learning are having an impact on most aspects of modern life, from entertainment, commerce, and healthcare. Netflix knows which films and series people prefer to watch, Amazon knows which items people like to buy when and where, and Google knows which symptoms and conditions people are searching for. All this data can be used for very detailed personal profiling, which may be of great value for behavioral understanding and targeting but also has potential for predicting healthcare trends. There is great optimism that the application of artificial intelligence (AI) can provide substantial improvements in all areas of healthcare from diagnostics to treatment. It is generally believed that AI tools will facilitate and enhance human work and not replace the work of physicians and other healthcare staff as such. AI is ready to support healthcare personnel with a variety of tasks from administrative workflow to clinical documentation and patient outreach as well as specialized support such as in image analysis, medical device automation, and patient monitoring. In this chapter, some of the major applications of AI in healthcare will be discussed covering both the applications that are directly associated with healthcare and those in the healthcare value chain such as drug development and ambient assisted living.

AI snapshot

Written from the paper’s full text.

TL;DR

This book chapter reviews major applications of artificial intelligence (AI) across healthcare, from precision medicine and drug discovery to imaging, wearables, robotics, and ambient assisted living. It argues that AI will generally augment rather than replace clinicians, but that adoption will be slowed by regulatory, integration, validation, and workflow challenges.

Key findings

  • The chapter reports that AI applications could cut annual US healthcare costs by USD 150 billion in 2026, and that the AI-associated healthcare market is expected to reach USD 6.6 billion by 2021 at a 40% compound annual growth rate.
  • It highlights performance examples including laparoscopic video analysis with 92.8% accuracy in identifying all procedure steps, AtomNet achieving an AUC of 0.9 or more for 58% of targets, and MoleculeNet having tested more than 700,000 compounds for toxicity or other properties.
  • It describes deep learning as using networks with more than 10 layers, compared with earlier neural networks of only 3–5 layers, and notes that OHDSI consolidated 1.26 billion patient records from 17 countries.
  • It states that half of people above age 65 have a disability, over 35 million people in the United States alone, and that AI is expected to support tasks such as administrative workflow, clinical documentation, image analysis, medical device automation, and patient monitoring.

Methods

  • Design: narrative book chapter/review, not a primary research study; it discusses selected applications of AI in healthcare and the healthcare value chain.
  • Data or sample: no primary dataset or study sample is analyzed; the chapter cites published literature, company examples, and prior studies, including a VR mastoidectomy study with one group of n=18 surgical trainees.
  • Main analysis: thematic discussion of precision medicine, genetics, drug discovery, medical visualization, AR/VR, personal health records, NLP, robotics, neuroprosthetics, ambient assisted living, and primary care AI.

Limitations

  • Stated: the chapter notes that telehealth solutions still need substantial independent validation for patient safety and efficacy, that IBM Watson has yet to show clinical value to patients, that evidence for gamification efficacy is scarce, and that surveyed primary care practitioners were skeptical about AI’s future role.
  • Stated: widespread adoption requires regulatory approval, EHR integration, standardization, clinician training, payer coverage, and field updates; the authors expect limited clinical use within 5 years and more extensive use within 10 years.
  • Evident: as a narrative review, it does not describe a systematic search strategy, inclusion criteria, quality appraisal, or quantitative synthesis, and many cited examples are early-stage, pilot, or company-reported rather than definitive clinical evidence.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/b978-0-12-818438-7.00002-2

Cites

59 of 59 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

13 of 75 cited sources checked · 7 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Telehealth technology is relevant in developing countries where the healthcare system is expanding and infrastructure can be designed to meet current needs.

    “Telehealth technology is also relevant in developing countries where the healthcare system is expanding and where healthcare infrastructure can be designed to meet the current needs [3].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [3] Combi, 2016, Telemedicine for developing countries, Appl Clin Inform source

      The Introduction says developing countries can benefit from telemedicine, and the Scalability section describes systems that can grow to serve more people and healthcare centers. The paper does not establish that developing countries’ healthcare systems are expanding or that their healthcare infrastructure can be designed to meet current needs; its Infrastructure section instead emphasizes existing barriers. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Premisecitation

    Drug discovery and development is an immensely long, costly, and complex process that can often take more than 10 years from identification of molecular targets until a drug product is approved and marketed.

    “Drug discovery and development is an immensely long, costly, and complex process that can often take more than 10 years from identification of molecular targets until a drug product is approved and marketed.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [14] Hughes, 2011, Principles of early drug discovery, Br J Pharmacol source
  • Premisecitation

    Blaschke et al. applied adversarial autoencoders and Bayesian optimization to generate ligands specific to the dopamine type 2 receptor.

    “Blaschke et al. also applied adversarial autoencoders and Bayesian optimization to generate ligands specific to the dopamine type 2 receptor [22].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [22] Blaschke, 2018, Application of generative autoencoder in de novo molecular design, Mol Inform source
  • Premisecitation

    Other biomarkers such as protein expression, gut microbiome, and metabolic profile are employed with machine learning to enable personalized treatments.

    “In addition to genetic information, other biomarkers such as protein expression, gut microbiome, and metabolic profile are also employed with machine learning to enable personalized treatments [12].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [12] Love-Koh, 2018, The future of precision medicine: potential impacts for health technology assessment, Pharmacoeconomics source

      In Findings > Technological Developments (page unavailable), the paper discusses proteomic and metabolomic biomarkers that may inform treatment choices, and AI-based algorithms that may recommend treatments using patient-level data. It does not identify gut microbiome biomarkers or say that the listed biomarkers are being used together with machine learning; it largely describes these developments as anticipated rather than established practice. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Premisecitation

    An assistive robot named RIBA with human-type arms was designed to help patients with lifting and moving heavy things.

    “An assistive robot named RIBA with human-type arms was designed to help patients with lifting and moving heavy things.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [57] Joseph A, Christian B, Abiodun AA, Oyawale F. A review on humanoid robotics in healthcare. In: MATEC Web of Conferences; 2018. https://www.matec-conferences.org/ source

      Section 4.3, “Humanoid Robot for Aging Population,” says RIBA had human-type arms and was designed for heavy physical tasks, specifically transferring a person between a bed and wheelchair. The “Author Robot” table likewise describes it as lifting and transferring patients. These passages do not support the claim that it helped patients lift or move heavy things. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
  • May combine two claims. 37
  • Classified as probably not a substantial claim. 35

  • May combine two claims.

    Premisecitation

    Mining of large-scale chemistry data is needed to efficiently classify potential drug compounds, and machine learning techniques have shown great potential.

    “However, mining of the large-scale chemistry data is needed to efficiently classify potential drug compounds and machine learning techniques have shown great potential [15].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [15] Ekins, 2019, Exploiting machine learning for end-to-end drug discovery and development, Nat Mater source

  • May combine two claims.

    Premisecitation

    Current AI applications within healthcare are often standalone applications used for diagnostics using medical imaging and disease prediction using remote patient monitoring.

    “Current AI applications within healthcare are often standalone applications, these are often used for diagnostics using medical imaging and for disease prediction using remote patient monitoring [38].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [38] Rojahn, 2016, Remote monitoring of chronic diseases: a landscape assessment of policies in four European countries, PLoS One source

      The Introduction (page not provided) says remote monitoring and trend analysis can detect deterioration in a patient’s condition early. It does not describe AI-based disease prediction, standalone AI applications, or diagnosis using medical imaging; its mention of transmitted still images is not evidence of AI imaging diagnostics. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    MoleculeNet is used to translate two-dimensional molecular structures into novel features/descriptors for predicting toxicity; more than 700,000 compounds have already been tested for toxicity or other properties.

    “Another platform called MoleculeNet is also used to translate two-dimensional molecular structures into novel features/descriptors, which can then be used in predicting toxicity of the given molecule. The MoleculeNet platform is built on data from various public databases and more than 700,000 compounds have already been tested for toxicity or other properties [20].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [20] Wu, 2018, MoleculeNet: a benchmark for molecular machine learning, Chem Sci source

  • May combine two claims.

    Premisecitation

    Merk et al. trained a recurrent neural network to capture bioactive compounds such as SMILES strings; identified compounds were synthesized and demonstrated potent receptor modulatory activity in in vitro assays.

    “Merk et al. trained a recurrent neural network to capture a large number of bioactive compounds such as SMILES strings. This model was then fine-tuned to recognize retinoid X and peroxisome proliferator-activated receptor agonists. The identified compounds were synthesized and demonstrated potent receptor modulatory activity in in vitro assays [23].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [23] Merk, 2018, De novo design of bioactive small molecules by artificial intelligence, Mol Inform source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    DeepTox, a DL-based model, has been implemented to assess toxicity of molecules based on a dataset containing many drug molecules.

    “Machine learning has also been implemented to assess the toxicity of molecules, for instance, using DeepTox, a DL-based model for evaluating the toxic effects of compounds based on a dataset containing many drug molecules [19].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [19] Mayr, 2016, DeepTox: toxicity prediction using deep learning, Front Environ Sci source

  • May combine two claims.

    Premisecitation

    In a cohort of diabetic and mental health patients, DeepCare was demonstrated to predict disease progression, optimal interventions, and likelihood for readmission.

    “Studying data from a cohort of diabetic and mental health patients it was demonstrated that DeepCare could predict the progression of disease, optimal interventions, and assessing the likelihood for readmission [37].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [37] Pham, 2017, Predicting healthcare trajectories from medical records: a deep learning approach, J Biomed Inform source

      The Conclusion reports that DeepCare was demonstrated on predicting next disease stages, recommending interventions, and estimating unplanned readmission among diabetic and mental health patients. But §4.3 evaluates prediction of interventions recorded for current diagnoses; it does not establish that the recommended interventions were optimal. (LLM, conf=0.89)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The AI-associated healthcare market is expected to grow rapidly and reach USD 6.6 billion by 2021, corresponding to a 40% compound annual growth rate.

    “The AI-associated healthcare market is expected to grow rapidly and reach USD 6.6 billion by 2021 corresponding to a 40% compound annual growth rate [4].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] Bresnick J. Artificial intelligence in healthcare market to see 40% CAGR surge; 2017.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Healthcare institutions struggle to keep up with new technological developments and high patient expectations.

    “Healthcare institutions are also fighting to keep up with all the new technological developments and the high expectations of patients with respect to levels of service and outcomes as they know it from consumer products including those of Amazon and Apple [2].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [2] Kirch, 2017, Addressing the physician shortage: the peril of ignoring demography, JAMA source

  • May combine two claims.

    Interpretationcitation

    For chronic stroke patients, immersive VR could act as a personal rehabilitation physiotherapist engaging upper limb movement multiple times a day, allowing possible neuroplasticity and gradual return of normal motor function.

    “For these patients, this immersive experience could act as a personal rehabilitation physiotherapist who engages their upper limb movement multiple times a day, allowing for possible neuroplasticity and a gradual return of normal motor function to these regions.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [31] Lee, 2020, Upper extremity rehabilitation using fully immersive virtual reality games with a head mount display: a feasibility study, PM R source

  • May combine two claims.

    Premisecitation

    NLP can provide substantial benefits in healthcare, including efficient billing, authorization approval, clinical decision support, and medical policy assessment.

    “There are many areas in healthcare in which NLP can provide substantial benefits. Some of the more immediate applications include [36]”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [36] Rangasamy ASS, Nadenichek R, Rayasam M. Natural language processing in healthcare; 2018.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    A group at Alpen-Adria Universität Klagenfurt uses surgery videos as training material to identify a specific intervention made by the surgeon; the algorithm recognizes the likelihood of the intervention and the specific region in the body.

    “An example of this is a group at the Institute of Information Technology at the Alpen-Adria Universität Klagenfurt that uses surgery videos as training material in order to identify a specific intervention made by the surgeon. For example, when an act of dissection or cutting is performed on the patient’s tissues or organs, the algorithm recognizes the likelihood of the intervention as well as the”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [27] Petscharnig, 2018, Learning laparoscopic video shot classification for gynecological surgery, Multimed Tools Appl source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    druGAN used adversarial autoencoders to generate new molecular fingerprints and drug designs incorporating features such as solubility and absorption based on predefined anticancer drug properties; results suggest substantial improvement in efficiency in generating new drug designs with specific properties.

    “As an example, the program druGAN used adversarial autoencoders to generate new molecular fingerprints and drug designs incorporating features such as solubility and absorption based on predefined anticancer drug properties. These results suggest a substantial improvement in the efficiency in generating new drug designs with specific properties [21].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] Kadurin, 2017, druGAN: an advanced generative adversarial autoencoder model for de novo generation of new molecules with desired molecular properties in silico, Mol Pharm source

  • May combine two claims.

    Premisecitation

    Advantages of precision medicine include reduced healthcare costs, reduction in adverse drug response, and enhancing effectivity of drug action.

    “Advantages of precision medicine include reduced healthcare costs, reduction in adverse drug response, and enhancing effectivity of drug action [11].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [11] Konieczny L, Roterman I. Personalized precision medicine. Bio-Algorithms Med-Syst 2019; 15. source

  • May combine two claims.

    Premisecitation

    Molecular properties including octanol, solubility melting point, and biological activity can be evaluated using neural fingerprints as demonstrated by Coley et al. and others, and used to predict new features of drug molecules.

    “In this way, molecular properties including octanol, solubility melting point, and biological activity can be evaluated as demonstrated by Coley et al. and others and be used to predict new features of the drug molecules [18].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [18] Coley, 2017, Convolutional embedding of attributed molecular graphs for physical property prediction, J Chem Inf Model source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Gamification elements are used to motivate and drive users to reach their goals.

    “These elements are used to motivate and drive users to reach their goals [34].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [34] Johnson, 2016, Gamification for health and wellbeing: a systematic review of the literature, Internet Interv source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    Game-like technologies such as VR and AR could enhance and enrich the learning experience for future medical and health-related disciplines.

    “So game-like technologies such as VR and AR could enhance and enrich the learning experience for future medical and health-related disciplines [29].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [29] Chien CH, Chen CH, Jeng TS. An interactive augmented reality system for learning anatomy structure. In: Proceedings of the International MultiConference of Engineers and Computer Scientists 2010, IMECS 2010; 2010. http://www.iaeng.org/publication/IMECS2018/

      Not checked in the free run.

    • Rests on an assumption: The current educational system is limited, and this limitation is a hindrance for interactive disciplines such as medicine.

  • May combine two claims.

    Premisecitation

    A study has shown that late-stage adult cancer patients can use immersive VR with minimum physical discomfort and benefit from an enhanced relaxed state, entertainment, and distraction.

    “A study has shown that late-stage adult cancer patients can use this technology with minimum physical discomfort and in return benefit from an enhanced relaxed state, entertainment, and a much-needed distraction [32].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [32] Baños, 2013, A positive psychological intervention using virtual reality for patients with advanced cancer in a hospital setting: a pilot study to assess feasibility, Support Care Cancer

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Drug molecules and associated features used in in silico models are transformed into vector format; data include molecular descriptors, molecular fingerprints, SMILES strings, and grids for convolutional neural networks.

    “The drug molecules and the associated features used in the in silico models are transformed into vector format so they can be read by the learning systems. Generally, the data used here include molecular descriptors (e.g., physicochemical properties) and molecular fingerprints (molecular structure) as well as simplified molecular input line entry system (SMILES) strings and grids for convolutional”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [17] Lavecchia, 2019, Deep learning in drug discovery: opportunities, challenges and future prospects, Drug Discov Today source

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    Remote monitoring and early detection of disease could be immensely beneficial for those with chronic conditions and the elderly.

    “Remote monitoring and picking up on early signs of disease could be immensely beneficial for those who suffer from chronic conditions and the elderly.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationreasoning

    For real-life and precise execution, AR would be more advantageous in healthcare settings than VR.

    “For real-life and precise execution, AR would be more advantageous in healthcare settings.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    The model sees sequences of words/speaking style and decides whether these emerging patterns are likely to be seen in individuals with and without depression.

    “According to the researchers, the “model sees sequences of words/speaking style” and decides whether these emerging patterns are likely to be seen in individuals with and without depression [63].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [63] Matheson R. Model can more naturally detect depression in conversations. MIT News; 2018.

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    AI used in healthcare can serve clinicians, patients, and other healthcare workers in four ways: assessment of disease onset and treatment success; management or alleviation of complications; patient-care assistance during a treatment or procedure; research aimed at discovery or treatment of disease.

    “A review by Becker [62] suggests that AI used in healthcare can serve clinicians, patients, and other healthcare workers in four different ways. Here, we will use these suggestions as inspirations and will expand on their contribution toward a successful implementation of AI in healthcare: (Fig. 2.9 )1.Assessment of disease onset and treatment success.2.Management or alleviation of complications.3”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [62] Becker, 2019, Artificial intelligence in medicine: what is it doing for us today?, Health Policy Technol source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Machine learning algorithms used in smart home applications include probabilistic and discriminative methods such as Naive Bayes classifier, Hidden Markov Model, support vector machine, and artificial neural networks.

    “Machine learning algorithms used in smart home applications include probabilistic and discriminative methods such as Naive Bayes classifier and Hidden Markov Model, support vector machine, and artificial neural networks [54].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [54] Fahad LG, Ali A, Rajarajan M. Learning models for activity recognition in smart homes. In: Information science and applications. Berlin: Springer; 2015. p. 819–26. source

  • May combine two claims.

    Premisecitation

    The artificial tactile sensing system can detect mass calcifications inside breast tissue based on palpation and comparison with reference data, and determine significant abnormalities.

    “The tactile sensory system can detect mass calcifications inside the breast tissue based on palpation of different points of the tissue and comparing with different reference data, and subsequently determine whether there are any significant abnormalities in the breast tissue.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [44] Naidu, 2017, A breakthrough in tumor localization: combining tactile sensing and ultrasound to improve tumor localization in robotics-assisted minimally invasive surgery, IEEE Robot Autom Mag source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The Kompaï robot used for the MARIO project was developed by Robosoft and contains a camera, a Kinect motion sensor, and two LiDAR remote sensing systems for navigation and object identification.

    “The Kompaï robot used for the MARIO project was developed by Robosoft and is a robot containing a camera, a Kinect motion sensor, and two LiDAR remote sensing systems for navigation and object identification [58].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [58] D’Onofrio, 2019, MARIO Project: validation and evidence of service robots for older people with dementia, J Alzheimers Dis source

  • May combine two claims.

    Premisecitation

    Assistive technologies increase the self-dependencies of patients, encouraging user participation in ICT tools to provide remote care services and information to healthcare professionals.

    “Assistive technologies increase the self-dependencies of patients, encouraging user participation in Information and Communication Technology (ICT) tools to provide remote care services type assistance and provide information to the healthcare professionals.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [51] Anderson, 2015, The impact of assistive technologies on formal and informal home care, Gerontologist source

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    Studies investigating cognitive stimulation seem to demonstrate a decrease in the rate of cognitive decline and progression of dementia.

    “Altogether, studies investigating cognitive stimulation seem to demonstrate a decrease in the rate of cognitive decline and progression of dementia.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Cognitive stimulation has been demonstrated to decrease cognitive impairment and can be trained using assistive robots.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Understanding microbiological niches (biomarkers) of trauma patients could hold the key to future wound infections and allow healthcare workers to prevent the worst outcome.

    “In a study published in the journal of trauma and acute care surgery, researchers think that by understanding the microbiological niches (biomarkers) of trauma patients, we could hold the key to future wound infections and therefore can allow healthcare workers to take the necessary arrangements to prevent the worst outcome [64].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [64] Dente, 2017, Towards precision medicine: accurate predictive modeling of infectious complications in combat casualties, J Trauma Acute Care Surg source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Virtrael is a cognitive stimulation platform that assesses, stimulates, and trains cognitive skills, based on visual memory training, with configuration, communication, and games functionalities.

    “Virtrael is one of such cognitive stimulation platforms and serves to assess, stimulate, and train various cognitive skills that experience a decline in the patient. The Virtrael program is based on visual memory training and the project is carried out by three different key functionalities: configuration, communication, and games.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [60] Garcia-Alonso J, Fonseca C, editors. Gerontechnology: First International Workshop. In: First international workshop on gerotechnology. Springer; 2018. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    OHDSI has consolidated 1.26 billion patient records from 17 different countries.

    “There are various international efforts in gathering EMR data across countries including Observational Health Data Science and Informatics (OHDSI), who have consolidated 1.26 billion patient records from 17 different countries [40].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [40] Wang, 2019, Deep learning in medicine—promise, progress, and challenges, JAMA Intern Med source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    AAL applications typically collect data through sensors and cameras and apply various artificially intelligent tools for developing an intelligent system.

    “AAL applications typically collect data through sensors and cameras and apply various artificially intelligent tools for developing an intelligent system [52].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [52] Barnay, 2016, Does home care for dependent elderly people improve their mental health?, J Health Econ source

  • May combine two claims.

    Premisecitation

    Most people want to preserve their autonomy even at an old age and maintain control over their lives and decisions.

    “Most people want to preserve their autonomy, even at an old age, and maintain control over their lives and decisions [50].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [50] Roberts AW, Ogunwole SU, Blakeslee L, Rabe MA, The population 65 years and older in the United States: 2016; 2018.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The games in Virtrael are intended to train cognitive skills including memory, attention, and planning.

    “The games are intended to train cognitive skills of the patient including memory, attention, and planning (Fig. 2.8 ) [60].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [60] Garcia-Alonso J, Fonseca C, editors. Gerontechnology: First International Workshop. In: First international workshop on gerotechnology. Springer; 2018. source

  • May combine two claims.

    Premisecitation

    DeepCare can model disease progression, support intervention recommendation, and provide disease prognosis based on EMR databases.

    “Using the stored data, the framework of DeepCare can model disease progression, support intervention recommendation, and provide disease prognosis based on EMR databases.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [41] Pham T, Tran T, Phung D, Venkatesh S. DeepCare: a deep dynamic memory model for predictive medicine. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer; 2016. source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    By daily collection of patient data, activities of daily living are defined over time and abnormalities can be detected as a deviation from the routine.

    “By daily collection of patient data, activities of daily living are defined over time and abnormalities can be detected as a deviation from the routine.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Low-cost sensors in an Internet of Things architecture can be a useful way of detecting abnormal behavior in the home.

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Integrating standalone AI applications with EMR data could provide greater value by adding personal medical data and history and a large statistical reference library to make classifications and predictions more accurate and powerful.

    “However, integrating such standalone applications with EMR data could provide even greater value by adding personal medical data and history as well as a large statistical reference library to make classifications and predictions more accurate and powerful.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    PARO is the most widely used robotic pet and carries sensors to sense touch, sounds, and visual objects.

    “The robotic pet PARO, a baby seal robot, is the most widely used robotic pet and carries various sensors to sense touch, sounds, and visual objects [61].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [61] Vitanza A, D’Onofrio G, Ricciardi F, Sancarlo D, Greco A, Giuliani F. Assistive robots for the elderly: innovative tools to gather health relevant data. In: Data science for healthcare: methodologies and applications. Springer; 2019. source

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Machine learning techniques can contribute toward the prediction of serious complications such as neuropathy in type 2 diabetes or early cardiovascular irregularities.

    “Machine learning techniques can also contribute toward the prediction of serious complications such as neuropathy that could arise for those suffering from type 2 diabetes or early cardiovascular irregularities.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationnone given

    Patient-care assistance technologies can improve the workflow for clinicians and contribute toward patient autonomy and well-being.

    “Patient-care assistance technologies can improve the workflow for clinicians and contribute toward patient’s autonomy and well-being.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    A main pain point is the lack of empathy and the ethical dilemma that can occur between AI and patients.

    “One main pain point refers to the lack of empathy and the ethical dilemma that can occur between AI and patients [70].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [70] Blease, 2019, Artificial intelligence and the future of primary care: exploratory qualitative study of UK general practitioners’ views, J Med Internet Res source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    During an epidemic, outbreak, natural or manmade disaster, or when the patient is away from their usual dwelling, technology that allows humans to remotely interact and solve problems will have to become a necessity.

    “It is also very important to note that at the time of an epidemic, an outbreak, natural or manmade disaster, or simply when the patient is away from their usual dwelling, a technology that allows humans to remotely interact and solve problems will have to become a necessity.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: At the time of writing (Early 2020), the threat of a SARS-COV-2 epidemic looms over many countries and is expanding at an unprecedented rate.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Speech and text recognition usage will increase.

    “Speech and text recognition are already employed for tasks like patient communication and capture of clinical notes, and their usage will increase.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    AI systems will not replace human clinicians on a large scale but will augment their efforts to care for patients.

    “It also seems increasingly clear that AI systems will not replace human clinicians on a large scale, but rather will augment their efforts to care for patients.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationreasoning

    The only healthcare providers who will risk their careers over time may be those who refuse to work alongside AI.

    “Perhaps the only healthcare providers who will risk their careers over time may be those who refuse to work alongside AI.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationnone given

    Widespread adoption of AI in healthcare requires approval by regulators, integration with EHR systems, standardization, training clinicians, payment by public or private payer organizations, and updates over time.

    “For widespread adoption to take place, AI systems must be approved by regulators, integrated with EHR systems, standardized to a sufficient degree that similar products work in a similar fashion, taught to clinicians, paid for by public or private payer organizations, and updated over time in the field.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The greatest challenge to AI in healthcare domains is ensuring adoption in daily clinical practice, not whether technologies will be capable enough.

  • May combine two claims.

    Interpretationreasoning

    These challenges will ultimately be overcome but will take much longer than it will take for the technologies themselves to mature.

    “These challenges will ultimately be overcome, but they will take much longer to do so than it will take for the technologies themselves to mature.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The greatest challenge to AI in healthcare domains is ensuring adoption in daily clinical practice, not whether technologies will be capable enough.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    If each patient is treated as an independent system, then based on the variety of designated data available, a bespoke approach can be implemented.

    “If each patient is treated as an independent system, then based on the variety of designated data available, a bespoke approach can be implemented.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Human clinicians may move toward tasks and job designs that draw on uniquely human skills like empathy, persuasion, and big-picture integration.

    “Over time, human clinicians may move toward tasks and job designs that draw on uniquely human skills like empathy, persuasion, and big-picture integration.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationreasoning

    This setting can be applied both at home and in a hospital setting to relieve work pressure from healthcare workers and improve service.

    “This setting can be applied both at home and in a hospital setting to relieve work pressure from healthcare workers and improve service.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    In a recent study, most practitioners were extremely skeptical regarding AI playing a significant role in the future of primary care, while acknowledging its potential benefits.

    “In a recent study, when asked about the future of AI on primary care, while acknowledging its potential benefits, most practitioners were extremely skeptical regarding it playing a significant role in the future of the profession.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [70] Blease, 2019, Artificial intelligence and the future of primary care: exploratory qualitative study of UK general practitioners’ views, J Med Internet Res source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Limited use of AI in clinical practice within 5 years and more extensive use within 10 years.

    “As a result, we expect to see limited use of AI in clinical practice within 5 years and more extensive use within 10 years.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    Patients could manage most of their own conditions at home and when necessary get in touch with a relevant healthcare worker who will refer them to more specialized physicians.

    “Combine this with the ever-increasing breakthroughs in the field of smart healthcare materials [71] and AI, one could envisage patients managing most of their own conditions at home and when necessary get in touch with a relevant healthcare worker who will refer them to more specialized physicians who could tend to their needs.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [71] Zang, 2015, Advances of flexible pressure sensors toward artificial intelligence and health care applications, Mater Horiz source
    • Rests on an assumption: Humanity prefers streamlining and creative solutions that are effective and take less out of our daily lives.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationreasoning

    Patients will be able to interact with devices remotely and access biometric data while feeling they are interacting with a caring and empathetic system.

    “Here, patients will be able to interact with the device in a remote manner and access their biometric data, all the while feeling that they are interacting with a caring and empathetic system that truly wants the best outcome for them.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationnone given

    AI will ultimately master diagnosis and treatment recommendations.

    “Although early efforts at providing diagnosis and treatment recommendations have proven challenging, we expect that AI will ultimately master that domain as well.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    AI can accelerate the diagnosis process and medical research.

    “AI can accelerate the diagnosis process and medical research.”
    Quote in the paperSubstantial claim
55 of 55

  • The paper relies on this without support.

    Demand for healthcare services is ever increasing and many countries are experiencing a shortage of healthcare practitioners, especially physicians.

    “The demand for healthcare services is ever increasing and many countries are experiencing a shortage of healthcare practitioners, especially physicians.”

  • The paper relies on this without support.

    AI applications can cut annual US healthcare costs by USD 150 billion in 2026.

    “It is estimated that AI applications can cut annual US healthcare costs by USD 150 billion in 2026.”

  • The paper relies on this without support.

    A large part of these cost reductions stem from changing the healthcare model from a reactive to a proactive approach, focusing on health management rather than disease treatment.

    “A large part of these cost reductions stem from changing the healthcare model from a reactive to a proactive approach, focusing on health management rather than disease treatment.”

  • The paper relies on this without support.

    This change is expected to result in fewer hospitalizations, less doctor visits, and less treatments.

    “This is expected to result in fewer hospitalizations, less doctor visits, and less treatments.”

  • The paper relies on this without support.

    There is a significant shortage of experts in medical image interpretation.

    “However, the demand is ever increasing and there is a significant shortage of experts in the field.”

  • The paper relies on this without support.

    Many inherited diseases result in symptoms without a specific diagnosis and interpreting whole genome data is still challenging due to the many genetic profiles.

    “Many inherited diseases result in symptoms without a specific diagnosis and while interpreting whole genome data is still challenging due to the many genetic profiles.”

  • The paper relies on this without support.

    There has been a substantial increase in the amount of data available assessing drug compound activity and biomedical data in the past few years.

    “There has been a substantial increase in the amount of data available assessing drug compound activity and biomedical data in the past few years.”

  • The paper relies on this without support.

    The smart tissue autonomous robot (STAR) can outperform human surgeons in some surgical procedures such as bowel anastomosis in animals.

    “The smart tissue autonomous robot (STAR) from the Johns Hopkins University has demonstrated that it can outperform human surgeons in some surgical procedures such as bowel anastomosis in animals.”

  • The paper relies on this without support.

    Video data is estimated to contain 25 times the amount of data from high-resolution diagnostic images such as CT.

    “Video data is estimated to contain 25 times the amount of data from high-resolution diagnostic images such as CT and could thus provide a higher data value based on resolution over time.”

  • The paper relies on this without support.

    The current educational system is limited, and this limitation is a hindrance for interactive disciplines such as medicine.

    “The current educational system is limited and for interactive disciplines such as medicine this can be a hindrance.”

  • The paper relies on this without support.

    A patient-centric personal health record promotes self-management and improves outcomes for patients.

    “However, in order to promote self-management and improve the outcomes for patients, a patient-centric personal health record should be implemented.”

  • The paper relies on this without support.

    Wearable health devices allow for constant measurement of certain vital signs under various conditions.

    “Wearable health devices (WHDs) are an upcoming technology that allow for constant measurement of certain vital signs under various conditions.”

  • The paper relies on this without support.

    Minimal invasive surgery requires different motor skills compared with conventional surgery due to lower tactile feedback.

    “Minimal invasive surgery requires different motor skills compared with conventional surgery due to the lower tactile feedback when relying more on tools and less on direct touching.”

  • The paper relies on this without support.

    Artificial tactile sensing offers advantages over physical touching, including a larger reference library, standardization among surgeons, continuous improvement, and level of training.

    “Artificial tactile sensing offers several advantages compared with physical touching including a larger reference library to compare sensation and standardization among surgeons with respect to quantitative features, continuous improvement, and level of training.”

  • The paper relies on this without support.

    Smart homes can be useful for people with dementia.

    “Smart homes can be useful for people with dementia and several studies have investigated smart home applications to facilitate the lives of dementia patients.”

  • The paper relies on this without support.

    Low-cost sensors in an Internet of Things architecture can be a useful way of detecting abnormal behavior in the home.

    “Low-cost sensors in an Internet of Things (IoT) architecture can be a useful way of detecting abnormal behavior in the home.”

  • The paper relies on this without support.

    Most robots currently under development have some level of AI technology incorporated for better performance.

    “Most of the robots currently under development have some level of AI technology incorporated for better performance with regard to classifications, language recognition, image processing, and more.”

  • The paper relies on this without support.

    There are more than a thousand EMR providers operating in the United States alone.

    “These use various operating systems and approaches with more than a thousand EMR providers operating in the United States alone.”

  • The paper relies on this without support.

    The only data often missing is data from in between interventions and between hospital visits when the patient is well or may not be showing symptoms.

    “The only data often missing is data from in between interventions and between hospital visits when the patient is well or may not be showing symptoms.”

  • The paper relies on this without support.

    Uncertainty modeling is important for monitoring patients with dementia because activities conducted by the patient are typically incomplete in nature.

    “Uncertainty modeling is important for monitoring patients with dementia as activities conducted by the patient are typically incomplete in nature.”

  • The paper relies on this without support.

    Domain knowledge related to the patient’s lifestyle combined with their medical history can enhance the probability of activity recognition and facilitate decision-making.

    “Domain knowledge related to the patient’s lifestyle is also important and combined with their medical history it can enhance the probability of activity recognition and facilitate decision-making.”

  • The paper relies on this without support.

    Many elderly people experience a decline in their cognitive abilities and have difficulties in problem-solving tasks as well as maintaining attention and accessing their memory.

    “Many elderly people experience a decline in their cognitive abilities and have difficulties in problem-solving tasks as well as maintaining attention and accessing their memory.”

  • The paper relies on this without support.

    Cognitive stimulation is a common rehabilitation approach after brain injuries from stroke, multiple sclerosis or trauma, and various mild cognitive impairments.

    “Cognitive stimulation is a common rehabilitation approach after brain injuries from stroke, multiple sclerosis or trauma, and various mild cognitive impairments.”

  • The paper relies on this without support.

    Cognitive stimulation has been demonstrated to decrease cognitive impairment and can be trained using assistive robots.

    “Cognitive stimulation has been demonstrated to decrease cognitive impairment and can be trained using assistive robots.”

  • The paper relies on this without support.

    Companion robots assist elderly patients with their stress or depression by connecting emotionally with the patient with enhanced social interaction and assistance with various daily tasks.

    “Such robots assist elderly patients with their stress or depression by connecting emotionally with the patient with enhanced social interaction and assistance with various daily tasks.”

  • The paper relies on this without support.

    AI is increasingly becoming an integral part of all our lives, from smartphones to cars and healthcare.

    “AI is increasingly becoming an integral part of all our lives. From smartphones to cars and more importantly our healthcare.”

  • The paper relies on this without support.

    The greatest challenge to AI in healthcare domains is ensuring adoption in daily clinical practice, not whether technologies will be capable enough.

    “The greatest challenge to AI in these healthcare domains is not whether the technologies will be capable enough to be useful, but rather ensuring their adoption in daily clinical practice.”

  • The paper relies on this without support.

    Prediction and assessment of a condition is something individuals will demand to have more control over in the coming years.

    “Prediction and assessment of a condition is something that individuals will demand to have more control over in the coming years.”

  • The paper relies on this without support.

    The general feeling of being unwell and its various complications that accompany mild illnesses are usually well tolerated by patients.

    “The general feeling of being unwell and its various complications that accompany mild illnesses are usually well tolerated by patients.”

  • The paper relies on this without support.

    The above scenario is the normal routine in most countries.

    “The above scenario is the normal routine in most countries.”

  • The paper relies on this without support.

    The current primary care system is not ideal and patients are likely to walk away dissatisfied.

    “However, despite the helpfulness of the physician, it is not an ideal system and it is likely that if you were in the position of the above patient, you will walk away dissatisfied with the care received.”

  • The paper relies on this without support.

    Frustration with such systems has led to an immense pressure on health workers and needs to be addressed.

    “The frustration with such systems has led to an immense pressure on the health workers and needs to be addressed.”

  • The paper relies on this without support.

    Trained AI bots can qualify whether certain symptoms warrant an actual conversation with a physician.

    “Prior to having access to an actual doctor, trained AI bots can qualify whether certain symptoms warrant an actual conversation with a physician.”

  • The paper relies on this without support.

    Questions and answers in AI triage are often vigorously reviewed by medical professionals at each stage to account for accuracy.

    “These questions and answers are often vigorously reviewed by medical professionals at each stage to account for accuracy.”

  • The paper relies on this without support.

    Remote digital visits allow patients to provide ample information for the physician to review before talking to the patient.

    “Normally, the patient books an appointment for a specific time, often during the same day. This provides them with ample time to provide as much information as possible for the physician responsible to review and carefully analyze the evidence before talking to the patient.”

  • The paper relies on this without support.

    Humanity prefers streamlining and creative solutions that are effective and take less out of our daily lives.

    “Humanity prefers streamlining and creative solutions that are effective and take less out of our daily lives.”

  • The paper relies on this without support.

    At the time of writing (Early 2020), the threat of a SARS-COV-2 epidemic looms over many countries and is expanding at an unprecedented rate.

    “At the time of writing (Early 2020), the threat of a SARS-COV-2 epidemic looms over many countries and is expanding at an unprecedented rate.”
37 of 37

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?

  • Achieving high predictive performance in healthcare often requires ML models with high flexibility to capture complex patterns in high-dimensional data.

    “Achieving high predictive performance in healthcare often requires ML models with high flexibility to capture complex patterns in high-dimensional data [41].”

    From Toward generalizable and interpretable machine learning models in healthcare: Insights from ICU outcome predictions (2026) · cites this paper as [41]

    The Precision medicine > Drug discovery and development section (page unavailable) describes ML prediction from large biomedical datasets and says deep-learning models often have many parameters. The Deep learning and medical image recognition section (page unavailable) says CNN layers learn image features and adjust weights to improve performance. Neither passage establishes that high predictive performance in healthcare often requires highly flexible models to capture complex patterns in high-dimensional data. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

  • Combining health data with demographic and environmental factors allows AI to make accurate predictions about future health outcomes and develop targeted interventions to prevent disease outbreaks.

    “By combining these data with other relevant information, such as demographic and environmental factors, AI can be used to make accurate predictions about future health outcomes and develop targeted interventions to prevent disease outbreaks [3,4].”

    From Edge AI for Early Detection of Chronic Diseases and the Spread of Infectious Diseases: Opportunities, Challenges, and Future Directions (2023) · cites this paper as [3]

  • AI-driven healthcare platforms can utilize facial recognition and voice analysis to detect emotions in patients during interactions.

    “AI-driven healthcare platforms can utilize facial recognition and voice analysis to detect emotions in patients during interactions [129].”

    From Ethical Framework for Harnessing the Power of AI in Healthcare and Beyond (2024) · cites this paper as [129]

  • Many other domains of medicine beyond transplantation could benefit from AI, including precision medicine, AI-assisted computer vision, augmented and virtual reality, and AI-assisted integration and collection of patients' records.

    “Moreover, while this manuscript focuses on the use of AI in transplantation, many other domains could benefit from it.”

    From Artificial Intelligence: Present and Future Potential for Solid Organ Transplantation (2022) · cites this paper as [61]

Cites

Flagged first, then by citations

2 of the 61 references couldn’t be matched to a known paper.

Research works printed but not linked: 2

  1. Kalis B, Collier M, Fu R. 10 promising AI applications in health care. Harvard Business Review; 2018.
  2. Davenport TH, Hongsermeier TM, Mc Cord KA. Using AI to improve electronic health records. Harvard Business Review; 2018.

Cited by

Sorted by citations

Related

Authors
Bohr, A., & Memarzadeh, K.
ORCID
Venue
Artificial Intelligence in Healthcare
Publisher
Elsevier
Published
2020
Pages
25-60
DOI
10.1016/b978-0-12-818438-7.00002-2
PMC
PMC7325854
OpenAlex
W3038004684
Semantic Scholar
fd41c526ab65ef21a43dfd2b427f8eaeea6c35ce
Cited by
1,636
References
61
Citations by year
2026: 180 · 2025: 400 · 2024: 557 · 2023: 327 · 2022: 114 · 2021: 56 · 2020: 1 · 2013: 1
FWCI
73.16
Type
book chapter
Language
en
Open access
Licence
Locations
  • Elsevier eBooks (publishedVersion)
  • PubMed Central (publishedVersion)
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Physical Sciences › Computer Science › Information Systems › Internet of Things and AI
Keywords
artificial intelligence in healthcare, machine learning, big data, clinical decision support, medical image analysis, medical device automation, patient monitoring, clinical documentation, administrative workflow, drug development, ambient assisted living, healthcare value chain

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.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Cited sources are checked by gpt-6-sol.

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