Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2024 Sep;11(36):e2400595.
doi: 10.1002/advs.202400595. Epub 2024 Jul 3.

Revolutionary Point-of-Care Wearable Diagnostics for Early Disease Detection and Biomarker Discovery through Intelligent Technologies

Fatemeh Haghayegh  1   2 , Alireza Norouziazad  1   2 , Elnaz Haghani  1   2 , Ariel Avraham Feygin  1   2 , Reza Hamed Rahimi  1   2 , Hamidreza Akbari Ghavamabadi  1   2 , Deniz Sadighbayan  3 , Faress Madhoun  1   2 , Manos Papagelis  2 , Tina Felfeli  4   5 , Razieh Salahandish  1   2
Affiliations
Review

Revolutionary Point-of-Care Wearable Diagnostics for Early Disease Detection and Biomarker Discovery through Intelligent Technologies

Fatemeh Haghayegh et al. Adv Sci (Weinh). 2024 Sep.

Abstract

Early-stage disease detection, particularly in Point-Of-Care (POC) wearable formats, assumes pivotal role in advancing healthcare services and precision-medicine. Public benefits of early detection extend beyond cost-effectively promoting healthcare outcomes, to also include reducing the risk of comorbid diseases. Technological advancements enabling POC biomarker recognition empower discovery of new markers for various health conditions. Integration of POC wearables for biomarker detection with intelligent frameworks represents ground-breaking innovations enabling automation of operations, conducting advanced large-scale data analysis, generating predictive models, and facilitating remote and guided clinical decision-making. These advancements substantially alleviate socioeconomic burdens, creating a paradigm shift in diagnostics, and revolutionizing medical assessments and technology development. This review explores critical topics and recent progress in development of 1) POC systems and wearable solutions for early disease detection and physiological monitoring, as well as 2) discussing current trends in adoption of smart technologies within clinical settings and in developing biological assays, and ultimately 3) exploring utilities of POC systems and smart platforms for biomarker discovery. Additionally, the review explores technology translation from research labs to broader applications. It also addresses associated risks, biases, and challenges of widespread Artificial Intelligence (AI) integration in diagnostics systems, while systematically outlining potential prospects, current challenges, and opportunities.

Keywords: AI‐assisted biomarker discovery; high‐throughput omics and clinical study; intelligent diagnostics; point‐of‐care biomarker detection platforms; precision medicine; remote disease detection; wearable health monitoring.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Schematic representation of the biomarker detection and identification platforms. Systems for the detection of various biomarkers are employed for disease diagnosis in a near‐patient format, in the presence of patients in clinical settings, or by other approaches such as physiological monitoring through wearable systems. Wearable platforms, such as smart contact lenses, pacemakers, sweat patches, and digital wristwatches provide invaluable insight into the physiological conditions and individual's health status. Biomarker detection and identification platforms for bedside, or near‐patient, applications are utilized for assessing the biocomposition of the patient biosamples and detecting specific biomarkers, or for medical imaging or medical signal analysis. Integrated with Artificial Intelligence (AI), upon transmission of data, preferably wirelessly, these systems can be used for a variety of purposes in the realm of biomarker recognition and characterization. This includes early detection of disease and Point‐Of‐Care (POC) applications for remote assessment of the individual's health status, with the testing performed by the patients outside clinical settings, conducting biological assays for biomolecule detection and biosample analysis, automated AI‐based image/signal analysis, omics for bio‐compositional analysis of biospecimens, and in the area of conducting clinical studies for biomarker discovery.
Figure 2
Figure 2
Examples of POC Testing (POCT) solutions for biomarker detection and In Vitro Diagnostics (IVDs). a) A POC detection platform for the detection of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) spike protein based on a self‐assembled plasmonic nanoprobe array mechanism enabled through Bioinspired Plasmo‐Virus (BPV) particle synthesis, Reproduced with permission.[ 40 ] Copyright 2023, American Chemical Society. b) Microfluidic system with integrated tapes for multiplexed detection of inflammatory markers of C‐reactive protein (CRP), Procalcitonin (PCT), and Interleukin‐6 (IL‐6) for early clinical diagnosis of sepsis, Reproduced with permission.[ 42 ] Copyright 2022, Elsevier. c) A dry chemistry‐based bipolar Electrochemiluminescence (ECL) immunoassay system for POCT of Alzheimer‐associated neuronal thread protein (AD7c‐NTP), Reproduced with permission.[ 47 ] Copyright 2023, American Chemical Society. d) GeneXpert platform, a PCR‐based system for the detection of multiple diseases in various testing settings (Image Reproduced with permission from Cepheid),[ 44 ] e) Mass screening of COVID‐19 via detection of breath volatile compounds using a SERS‐based breathalyzer, Reproduced with permission.[ 45 ] Copyright 2022, American Chemical Society. f) POC detection of COVID‐19 using electrochemical dual immunosensing of the nucleocapsid protein (N‐protein) via electrochemical impedance spectroscopy (Bi‐ECDAQ), Reproduced with permission.[ 58 ] Copyright 2022, Elsevier.
Figure 3
Figure 3
Microfluidics integrated biosensors. a) Immunobiosensor on a chip for electrochemical detection of COVID‐19 in a self‐powered microfluidic system, Reproduced with permission.[ 63 ] Copyright 2022, Royal Society of Chemistry. b) Multiplexed DNA detection for Malaria diagnosis using an origami‐based paper microfluidic system and lateral flow detection, Reproduced with permission.[ 64 ] Copyright 2019, PNAS. c) Paper‐based sample concentration technique in the Lateral Flow Assay (LFA) for detection of Human Immunodeficiency Virus (HIV) nucleic acid, Reproduced with permission.[ 65 ] Copyright 2016, Elsevier. d) Programmed passive microfluidic capillary system and embedded Enzyme‐Linked Immunosorbent Assay (ELISA) for detection of SARS‐CoV‐2 antibodies, Reproduced with permission.[ 66 ] Copyright 2022, Springer Nature. e) Active microfluidic system for rapid detection of cardiac troponin I (cTnI) through on‐chip ELISA, Reproduced with permission.[ 67 ] Copyright 2021, Springer Nature. f) Droplet microfluidic system and integrated Surface‐Enhanced Raman spectroscopy (SERS)‐based sensor, Reproduced with permission.[ 68 ] Copyright 2019, Royal Society of Chemistry. g) Autonomous microfluidic system coupled with electrochemical immunosensor for detection of Glial Fibrillary Acidic Proteins (GFAP), Reproduced with permission.[ 69 ] Copyright 2022, Royal Society of Chemistry. h) Detection of Staphylococcus aureus DNA via a low‐cost microfluidic integrated isothermal amplification and on‐site nucleic acid quantification; (SIMPLE: Self‐powered Integrated Microfluidic Point‐of‐care Low‐cost Enabling) chip, Reproduced with permission.[ 70 ] Copyright 2017, Science.
Figure 4
Figure 4
Illustration of wearable systems and POCT. Wearable technology platforms are currently being extensively investigated for their capability to detect biomarkers in various biofluids for POCT and on‐site operation by patients. This includes the use of wearable sweat collection/detection patches for sweat analysis, Reproduced with permission.[ 85 ] Copyright 2021, Advanced Science published by Wiley‐VCH GmbH), dermal patches for Interstitial Fluid (ISF) analysis, Reproduced with permission.[ 20b ] Copyright 2022, Springer Nature Limited, oral wearables for saliva analysis, smart contact lenses for tear analysis, and smartwatch systems for physiological monitoring. The detection methodologies employed in these platforms encompass electrochemical, colorimetry, and optical transduction, with various readout modules such as potentiostat systems or optical light intensity detection, all techniques enabling miniaturized POCT. These point‐of‐care solutions are versatile in terms of their microfabrication technique, potential for integration with smart systems, as well as applicability for the detection of a panel of biomarkers through multiplexing.
Figure 5
Figure 5
Wearable systems for biomarker detection and on‐site health monitoring. A, a) Colorimetric sweat chloride analysis enabled through skin mount microfluidic system featuring superabsorbent polymer valves, Reproduced with permission.[ 114 ] Copyright 2018, John Wiley and Sons. b) Miniaturized microfluidics and colorimetric analysis for detection of nutrients in sweat and supplying vitamins, Reproduced with permission.[ 85 ] Copyright 2021, Wiley‐VCH, c) The Gx sweat patch for personalized sweat rate determination and sweat chloride analysis for athletic use, Reproduced with permission.[ 115 ] Copyright 2023, Springer Nature Limited. d) A sweat patch embedded with electrochemical sensors for C‐reactive protein (CRP) monitoring in sweat, Reproduced with permission.[ 103 ] Copyright 2023, Springer Nature Limited, e) MicroSweat: A capillary microfluidic sweat collection patch for stress monitoring via determination of sweat cortisol levels, Reproduced with permissions.[ 86 ] Copyright 2022, Wiley‐VCH. f) Electrochemical urine analysis via a wearable diaper sensor for urinary incontinence complications, Reproduced with permission.[ 94 ] Copyright 2022, Elsevier. g) Wearable Electroencephalogram (EEG) device with a Brain–AI Closed‐Loop System (BACLoS) for predicting human cognitive consequences, Reproduced with permission.[ 116 ] Copyright 2022, Springer Nature Limited. h) Microfluidics skin interface sweat analysis with a 3D complex structure enabling integration of colorimetric assays evaluating sweat chloride, Reproduced with permission.[ 105 ] Copyright 2016, The American Association for the Advancement of Science. i) Respiration sensor in wearable format inside a mask for chronic kidney disease monitoring through ammonia (NH3) content measurement, Reproduced with permission.[ 95 ] Copyright 2022, American Chemical Society. j) Soft wearable microfluidics for sweat capture, storage, and analysis using the smartphone‐assisted colorimetric technique, Reproduced with permission.[ 106 ] Copyright 2016, The American Association for the Advancement of Science. B) The procedure of data analysis using AI algorithms from data collection, preprocessing, and dataset split to test and train, training the Machine Learning (ML) or Deep Learning (DL) model, validating the model, and final testing for predictions. C) The output of the wearable systems integrated with AI algorithms can be represented as clustering the data, determining the true or false negative/positive rates, and the accuracy of the model and analysis.
Figure 6
Figure 6
The role of AI in technologies for disease detection. AI systems are widely incorporated in the analysis of medical signals, as well as medical images, along with the design and analysis of bioassays and POC platforms. The methodology for analyzing systems using AI solutions, such as ML and DL methods involves data pre‐processing, labeling, selecting appropriate parameters, categorization of data, selecting the proper architecture, training the model, and testing and validation.
Figure 7
Figure 7
AI and disease detection through clinical and medical data, electronic health records, and medical libraries. The analysis involves examining medical images such as computed tomography (CT)‐ scans, for example for ML‐based examination of the Thoracic Aorta,[ 155 ] Magnetic Resonance Imaging (MRI), for instance, to determine lung diseases,[ 140b ] ultrasounds,[ 141 ] or Optical Coherence Tomography (OCT) imaging.[ 156 ] Medical signals including EEG,[ 157 ] Electrocardiogram (ECG),[ 146 ] or Electromyography (EMG)[ 147 ] can also be used as inputs of the AI algorithms. Evaluating the electronic medical health records such as physician data, hospital treatment records, biometric data, thesis, patient information, and insurance information, along with medical libraries including the Cancer Imaging Archive (TCIA),[ 148 ] the Medical Information Mart for Intensive Care (MIMIC‐III) dataset,[ 149 ] Physionet EEG Motor Movement/Imagery Dataset,[ 143 ] and DEAP: A Database for Emotion Analysis using Physiological Signals,[ 152 ] for AI‐based disease detection and prediction.
Figure 8
Figure 8
Commercialization of AI‐integrated diagnostic tools. Translation of the laboratory research products to the market, with examples including the AliveCor Mobile ECG system (Image reproduced with permission from AliveCor Inc.),[ 161 ] and the Butterfly portable ultrasound system (Icon of Butterfly ultrasound probe provided with permission from Butterfly Network, Inc.).[ 160 ] For this purpose, one should consider factors such as pathways for regulatory approvals, approaches for attracting investments, the possible bias in the AI algorithms that affect the security, transparency, and reliability of the data analysis, and interactions between the clinicians, patients, and other customers in the market.
Figure 9
Figure 9
Bioassays and AI. Biological assays including microfluidic assays, in vitro toxicity assays, or immunoassays, as well as protein quantification such as western blotting can be integrated with statistical/ML methods for data analysis, including regression approaches such as Quantitative Structure‐Activity Relationship (QSAR), k‐means clustering, and k‐nearest neighbours, ensemble forest methods, Support Vector Machine (SVM), and Principal Component Analysis (PCA) with applications in enhanced cell tracking, Reproduced with permission.[ 229 ] Copyright 2023, Frontiers, image processing for paper‐based microfluidics, Reproduced with permission.[ 232 ] Copyright 2021, American Chemical Society, and protein classification/generation.
Figure 10
Figure 10
Integration of microfluidics in the POC systems and platforms for conducting biological assay with AI technologies. A) Various microfluidics elements such as microchannels, valves, and droplet microfluidic systems that can be used for developing platforms to conduct biological assays. Such techniques coupled with sensing modalities enable on‐site biosample analysis with marketed and FDA‐authorized examples including the Maverick™ SARS‐CoV‐2 Multi‐Antigen Serology Panel [ 244 ](Image reproduced with permission from Genalyte Inc.), or the Minuteful smartphone‐powered kidney test (Image reproduced with permission from Healthy.io Ltd).[ 245 ] The role of AI systems in biological assays and clinical decision‐making is also becoming more established with marketed technologies such as EasyScan One (the newer version of EasyScan Go), a microscopy system for Malaria detection based on machine learning [ 246 ](Image reproduced with permission from Motic Instruments Inc), and the Tempus one system, one of the latest innovations of the Tempus labs incorporating generative AI solutions in precision medicine (Image reproduced with permission from Tempus).[ 243 ] Recent research works explore novel applications of the integration of AI systems into the biological assays and biomarker detection systems, including a) AI‐assisted urinary multimarker sensor for prostate cancer screening, Reproduced with permission.[ 203 ] Copyright 2021, American Chemical Society. b) Electrochemiluminescence biosensor with smart‐phone integrated and machine‐learning assisted algorithm for detection of various metabolites, Reproduced with permission.[ 208 ] Copyright 2023, Elsevier. c) A microfluidic digital immunoassay for inflammatory markers and antibiotics detection empowered by a computer vision‐based AI‐mediated encoding‐decoding system, Reproduced with permission.[ 209 ] Copyright 2023, American Chemical Society. d) Utility of the artificial neural network in processing the light parameters of the fluorescence for optical POC solutions, Reproduced with permission.[ 216 ] Copyright 2023, Elsevier. e) High‐throughput SERS‐based classification of the cell secretomes assisted by the machine learning algorithms, Reproduced with permission.[ 217 ] Copyright 2023, Wiley VCH.
Figure 11
Figure 11
Biomarker discovery via wearable systems, microscopy and medical image analysis, signal analysis, omics, and Electronic Health Records (EHR). Clinical studies are conducted for collecting bio‐physiological data, either in a wearable format (Reproduced with permission.[ 20b ] Copyright 2022, Springer Nature Limited) or via imaging, microscopy, or medical signal acquisition. The acquired data include pathological microscopy images requiring segmentation/labeling (Reproduced with permission.[ 250 ] Copyright 2012, American Association of Neuropathologists, Inc.) medical signals needing feature extraction as specific disease markers, radiology images for anomaly discovery, or biocompositional data analysis such as the outputs of the omics assessments including Liquid Chromatography‐Mass Spectroscopy (LC‐MS) or proteomics heatmaps. The EHRs such as data from hospitals, patients, lab tests, and radiology reports can also be used for introducing specific disease biomarkers in this realm upon analysis of collected data.
Figure 12
Figure 12
Omics and Biomarker Discovery. Biocompositional analysis of various biosamples including blood, saliva, tear, urine, and sweat through genomics, proteomics, metabolomics, transcriptomics, and epigenetics leads to the introduction of novel disease markers upon analyzing the generated chromatography spectra, transcriptomics heatmaps, gene expressions, and proteome patterns. This includes AI, ML, or DL analysis, extracting features associated with disease‐specific biomolecules, extracting correlations and significant differences in intensities/concentrations, and analysis of accuracy.
Figure 13
Figure 13
Utilization of POC systems and wearable platforms for biomarker discovery. Moving from conventional biomolecular and biocompositional analysis methods for biomarker discovery in clinical studies, such as omics approaches, LC‐MS, or ELISA, POC biomarker detection systems can be used in a high‐throughput format, remotely analyzing biosamples or medical image/data for extracting patterns and discovering new features/markers. AI methods can play a pivotal role in introducing reliable disease‐specific biomarkers by analyzing the collected data from multiple sources including wearable systems, medical data, medical images, medical signals, and multiplexed POCT systems, and then drawing on correlations and associations.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Balogh E. P., Miller B. T., Ball J. R., Improving Diagnosis in Health Care, National Academies Press (US), Washington (DC) 2015. - PubMed
    1. Black G. B., Lyratzopoulos G., Vincent C. A., Fulop N. J., Nicholson B. D., BMJ 2023, 380, 071225.
    1. a) Califf R. M., Exp. Biol. Med. 2018, 243, 213; - PMC - PubMed
    2. b) Chiu F.‐Y., Yen Y., Biomarker Res. 2023, 11, 35. - PMC - PubMed
    1. Ahmad A., Imran M., Ahsan H., Pharmaceutics 2023, 15, 1630. - PMC - PubMed
    1. Vinood P., Victor P., Textbook of Cancer Epidemiology, Springer, Dordrecht, The Netherlands: 2015.

LinkOut - more resources