Submit to Special Issue Submit Abstract to Special Issue Review for JCM Propose a Special Issue

Journal Browser

The Convergence of Human and Artificial Intelligence on Clinical Care - Part II

Special Issue Editors
Special Issue Information
Keywords
Published Papers

A special issue of Journal of Clinical Medicine (ISSN 2077-0383). This special issue belongs to the section "Epidemiology & Public Health".

Deadline for manuscript submissions: 15 July 2024 | Viewed by 33981

Share This Special Issue

Special Issue Editor

Dr. Vida Abedi

E-Mail Website
Guest Editor

Department of Public Health Science, Penn State College of Medicine, Hershey, PA, USA
Interests: machine learning; clinical informatics; healthcare innovation; EHR/EMR mining; natural language processing; complex diseases; outcome prediction; health disparity; machine learning-enabled decision support system; stroke; transient ischemic attack; cerebrovascular medicine
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Given the enormous success of the Part I issue (The Convergence of Human and Artificial Intelligence on Clinical Care - Part I), I am thrilled to move forward to Special Issue Part II. Briefly, Part I was a successful collection not only because of the quality of the articles but also because of the range of AI applications in healthcare. Briefly, Part I of the collection contains 12 studies, large and pilots, that try to tackle the changing landscape of healthcare using AI. The articles are in five main areas: (i) using adaptive imputation to increase the density of clinical data for improving downstream modeling, (ii) machine learning-empowered diagnosis models, (iii) machine learning models for outcome prediction, (iv) innovative use of AI to improve our understanding of the public view, and (v) understanding of the attitude of providers in trusting insights from AI for complex cases. Overall, studies used an array of data modalities, including data from electronic health records, imaging data, voice signals, resource utilization, Twitter data, and questionnaire, in addition to a wide range of modeling frameworks, designs, and algorithms.

In Part II, we are focusing not only on methodological advances for diverse data types but also on the implementation component. We particularly welcome articles providing new insights into (i) the ethical and technological challenges when integrating AI into the clinical workflow; (ii) effectiveness and clinical value of tools and AI-enabled decision support systems for improving care; (iii) different ways that AI can be used to improve access, reduce health disparity, and improve outcome; and finally, (iv) other application of AI in pre-clinical and clinical settings, including but not limited to leveraging AI for improving clinical trials among others. We welcome both solicited and unsolicited submissions that will contribute to this goal.

Dr. Vida Abedi
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Journal of Clinical Medicine is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

Precision medicine
Digitalization of health care, challenges, and opportunities
Implementation and adoption of novel technologies in healthcare
Patient stratification and subtyping
Personalized care management
Machine learning-enabled decision support system
Providers-in-the-loop in the era of AI
Improving diagnosis accuracy
EHR/EMR mining
Optimization models for shared decision making in healthcare

Published Papers (11 papers)

Download All Papers

Research

Jump to: Review, Other

13 pages, 1840 KiB

Open AccessArticle

Prediction of Blood Risk Score in Diabetes Using Deep Neural Networks

by J. Quetzalcóatl Toledo-Marín, Taqdir Ali, Tibor van Rooij, Matthias Görges and Wyeth W. Wasserman

J. Clin. Med. 2023, 12(4), 1695; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm12041695 - 20 Feb 2023

Viewed by 1617

Abstract

Improving the prediction of blood glucose concentration may improve the quality of life of people living with type 1 diabetes by enabling them to better manage their care. Given the anticipated benefits of such a prediction, numerous methods have been proposed. Rather than attempting to predict glucose concentration, a deep learning framework for prediction is proposed in which prediction is performed using a scale for hypo- and hyper-glycemia risk. Using the blood glucose risk score formula proposed by Kovatchev et al., models with different architectures were trained, including, a recurrent neural network (RNN), a gated recurrent unit (GRU), a long short-term memory (LSTM) network, and an encoder-like convolutional neural network (CNN). The models were trained using the OpenAPS Data Commons data set, comprising 139 individuals, each with tens of thousands of continuous glucose monitor (CGM) data points. The training set was composed of 7% of the data set, while the remaining was used for testing. Performance comparisons between the different architectures are presented and discussed. To evaluate these predictions, performance results are compared with the last measurement (LM) prediction, through a sample-and-hold approach continuing the last known measurement forward. The results obtained are competitive when compared to other deep learning methods. A root mean squared error (RMSE) of 16 mg/dL, 24 mg/dL, and 37 mg/dL were obtained for CNN prediction horizons of 15, 30, and 60 min, respectively. However, no significant improvements were found for the deep learning models compared to LM prediction. Performance was found to be highly dependent on architecture and the prediction horizon. Lastly, a metric to assess model performance by weighing each prediction point error with the corresponding blood glucose risk score is proposed. Two main conclusions are drawn. Firstly, going forward, there is a need to benchmark model performance using LM prediction to enable the comparison between results obtained from different data sets. Secondly, model-agnostic data-driven deep learning models may only be meaningful when combined with mechanistic physiological models; here, it is argued that neural ordinary differential equations may combine the best of both approaches. These findings are based on the OpenAPS Data Commons data set and are to be validated in other independent data sets. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

12 pages, 1130 KiB

Open AccessArticle

A Simpler Machine Learning Model for Acute Kidney Injury Risk Stratification in Hospitalized Patients

by Yirui Hu, Kunpeng Liu, Kevin Ho, David Riviello, Jason Brown, Alex R. Chang, Gurmukteshwar Singh and H. Lester Kirchner

J. Clin. Med. 2022, 11(19), 5688; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11195688 - 26 Sep 2022

Cited by 4 | Viewed by 1637

Abstract

Background: Hospitalization-associated acute kidney injury (AKI), affecting one-in-five inpatients, is associated with increased mortality and major adverse cardiac/kidney endpoints. Early AKI risk stratification may enable closer monitoring and prevention. Given the complexity and resource utilization of existing machine learning models, we aimed to develop a simpler prediction model. Methods: Models were trained and validated to predict risk of AKI using electronic health record (EHR) data available at 24 h of inpatient admission. Input variables included demographics, laboratory values, medications, and comorbidities. Missing values were imputed using multiple imputation by chained equations. Results: 26,410 of 209,300 (12.6%) inpatients developed AKI during admission between 13 July 2012 and 11 July 2018. The area under the receiver operating characteristic curve (AUROC) was 0.86 for Random Forest and 0.85 for LASSO. Based on Youden’s Index, a probability cutoff of >0.15 provided sensitivity and specificity of 0.80 and 0.79, respectively. AKI risk could be successfully predicted in 91% patients who required dialysis. The model predicted AKI an average of 2.3 days before it developed. Conclusions: The proposed simpler machine learning model utilizing data available at 24 h of admission is promising for early AKI risk stratification. It requires external validation and evaluation of effects of risk prediction on clinician behavior and patient outcomes. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

16 pages, 1908 KiB

Open AccessArticle

MobileSkin: Classification of Skin Lesion Images Acquired Using Mobile Phone-Attached Hand-Held Dermoscopes

by Abdurrahim Yilmaz, Gulsum Gencoglan, Rahmetullah Varol, Ali Anil Demircali, Meysam Keshavarz and Huseyin Uvet

J. Clin. Med. 2022, 11(17), 5102; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11175102 - 30 Aug 2022

Cited by 6 | Viewed by 3735

Abstract

Dermoscopy is the visual examination of the skin under a polarized or non-polarized light source. By using dermoscopic equipment, many lesion patterns that are invisible under visible light can be clearly distinguished. Thus, more accurate decisions can be made regarding the treatment of skin lesions. The use of images collected from a dermoscope has both increased the performance of human examiners and allowed the development of deep learning models. The availability of large-scale dermoscopic datasets has allowed the development of deep learning models that can classify skin lesions with high accuracy. However, most dermoscopic datasets contain images that were collected from digital dermoscopic devices, as these devices are frequently used for clinical examination. However, dermatologists also often use non-digital hand-held (optomechanical) dermoscopes. This study presents a dataset consisting of dermoscopic images taken using a mobile phone-attached hand-held dermoscope. Four deep learning models based on the MobileNetV1, MobileNetV2, NASNetMobile, and Xception architectures have been developed to classify eight different lesion types using this dataset. The number of images in the dataset was increased with different data augmentation methods. The models were initialized with weights that were pre-trained on the ImageNet dataset, and then they were further fine-tuned using the presented dataset. The most successful models on the unseen test data, MobileNetV2 and Xception, had performances of 89.18% and 89.64%. The results were evaluated with the 5-fold cross-validation method and compared. Our method allows for automated examination of dermoscopic images taken with mobile phone-attached hand-held dermoscopes. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

15 pages, 2642 KiB

Open AccessArticle

Prediction of Influenza Complications: Development and Validation of a Machine Learning Prediction Model to Improve and Expand the Identification of Vaccine-Hesitant Patients at Risk of Severe Influenza Complications

by Donna M. Wolk, Alon Lanyado, Ann Marie Tice, Maheen Shermohammed, Yaron Kinar, Amir Goren, Christopher F. Chabris, Michelle N. Meyer, Avi Shoshan and Vida Abedi

J. Clin. Med. 2022, 11(15), 4342; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11154342 - 26 Jul 2022

Cited by 4 | Viewed by 1918

Abstract

Influenza vaccinations are recommended for high-risk individuals, but few population-based strategies exist to identify individual risks. Patient-level data from unvaccinated individuals, stratified into retrospective cases (n = 111,022) and controls (n = 2,207,714), informed a machine learning model designed to create an influenza risk score; the model was called the Geisinger Flu-Complications Flag (GFlu-CxFlag). The flag was created and validated on a cohort of 604,389 unique individuals. Risk scores were generated for influenza cases; the complication rate for individuals without influenza was estimated to adjust for unrelated complications. Shapley values were used to examine the model’s correctness and demonstrate its dependence on different features. Bias was assessed for race and sex. Inverse propensity weighting was used in the derivation stage to correct for biases. The GFlu-CxFlag model was compared to the pre-existing Medial EarlySign Flu Algomarker and existing risk guidelines that describe high-risk patients who would benefit from influenza vaccination. The GFlu-CxFlag outperformed other traditional risk-based models; the area under curve (AUC) was 0.786 [0.783–0.789], compared with 0.694 [0.690–0.698] (p-value < 0.00001). The presence of acute and chronic respiratory diseases, age, and previous emergency department visits contributed most to the GFlu-CxFlag model’s prediction. When higher numerical scores were assigned to more severe complications, the GFlu-CxFlag AUC increased to 0.828 [0.823–0.833], with excellent discrimination in the final model used to perform the risk stratification of the population. The GFlu-CxFlag can better identify high-risk individuals than existing models based on vaccination guidelines, thus creating a population-based risk stratification for individual risk assessment and deployment in vaccine hesitancy reduction programs in our health system. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

13 pages, 652 KiB

Open AccessArticle

Neural Network Aided Detection of Huntington Disease

by Gerardo Alfonso Perez and Javier Caballero Villarraso

J. Clin. Med. 2022, 11(8), 2110; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11082110 - 10 Apr 2022

Cited by 4 | Viewed by 1752

Abstract

Huntington Disease (HD) is a degenerative neurological disease that causes a significant impact on the quality of life of the patient and eventually death. In this paper we present an approach to create a biomarker using as an input DNA CpG methylation data to identify HD patients. DNA CpG methylation is a well-known epigenetic marker for disease state. Technological advances have made it possible to quickly analyze hundreds of thousands of CpGs. This large amount of information might introduce noise as potentially not all DNA CpG methylation levels will be related to the presence of the illness. In this paper, we were able to reduce the number of CpGs considered from hundreds of thousands to 237 using a non-linear approach. It will be shown that using only these 237 CpGs and non-linear techniques such as artificial neural networks makes it possible to accurately differentiate between control and HD patients. An underlying assumption in this paper is that there are no indications suggesting that the process is linear and therefore non-linear techniques, such as artificial neural networks, are a valid tool to analyze this complex disease. The proposed approach is able to accurately distinguish between control and HD patients using DNA CpG methylation data as an input and non-linear forecasting techniques. It should be noted that the dataset analyzed is relatively small. However, the results seem relatively consistent and the analysis can be repeated with larger data-sets as they become available. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

22 pages, 5756 KiB

Open AccessArticle

The Computer Simulation of Therapy with the NMDA Antagonist in Excitotoxic Neurodegeneration in an Alzheimer’s Disease-like Pathology

by Dariusz Świetlik, Aida Kusiak, Marta Krasny and Jacek Białowąs

J. Clin. Med. 2022, 11(7), 1858; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11071858 - 27 Mar 2022

Cited by 7 | Viewed by 2143

Abstract

(1) Background: The use of uncompetitive N-methyl-D-aspartate (NMDA) receptor antagonists results in neuroprotective benefits in patients with moderate to severe Alzheimer’s disease. In this study, we demonstrated mathematical and computer modelling of the excitotoxicity phenomenon and performed virtual memantine therapy. (2) Methods: A computer simulation environment of the N-methyl-D-aspartate receptor combining biological mechanisms of channel activation by means of excessive extracellular glutamic acid concentration in three models of excitotoxicity severity. The simulation model is based on sliding register tables, where each table is associated with corresponding synaptic inputs. Modelling of the increase in extracellular glutamate concentration, through over-stimulation of NMDA receptors and exacerbation of excitotoxicity, is performed by gradually increasing the parameters of phenomenological events by the power function. Pathological models were virtually treated with 3–30 µM doses of memantine compared to controls. (3) Results: The virtual therapy results of memantine at doses of 3–30 µM in the pathological models of excitotoxicity severity show statistically significant neuroprotective benefits in AD patients with moderate severity, 1.25 (95% CI, 1.18–1.32) vs. 1.76 (95% CI, 1.71–1.80) vs. 1.53 (95% CI, 1.48–1.59), (p < 0.001), to severe, 1.32 (95% CI, 1.12–1.53) vs. 1.77 (95% CI, 1.72–1.82) vs. 1.73 (95% CI, 1.68–1.79), (p < 0.001), in the area of effects on memory. A statistically significant benefit of memantine was demonstrated for all neuronal parameters in pathological models. In the mild severity model, a statistically significant increase in frequency was obtained relative to virtual memantine treatment with a dose of 3 µM, which was 23.5 Hz (95% CI, 15.5–28.4) vs. 38.8 Hz (95% CI, 34.0–43.6), (p < 0.0001). In the intermediate excitotoxicity severity model, a statistically significant increase in frequency was obtained relative to virtual memantine therapy with a 3 µM dose of 26.0 Hz (95% CI, 15.7–36.2) vs. 39.0 Hz (95% CI, 34.2–43.8) and a 10 µM dose of 26.0 Hz (95% CI, 15.7–36.2) vs. 30.9 Hz (95% CI, 26.4–35.4), (p < 0.0001). A statistically significant increase in frequency was obtained in the advanced excitotoxicity severity model as in the medium. (4) Conclusions: The NMDA antagonist memantine causes neuroprotective benefits in patients with moderate to severe AD. One of the most important benefits of memantine is the improvement of cognitive function and beneficial effects on memory. On the other hand, memantine provides only symptomatic and temporary support for AD patients. Memantine is prescribed in the US and Europe if a patient has moderate to severe AD. Memantine has also been approved for mild to moderate AD patients. However, its very modest effect provides motivation for further research into new drugs in AD. We are the first to present a mathematical model of the NMDA receptor that allows the simulation of excitotoxicity and virtual memantine therapy. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

Review

Jump to: Research, Other

32 pages, 17043 KiB

Open AccessFeature PaperReview

Digital Therapeutics for Improving Effectiveness of Pharmaceutical Drugs and Biological Products: Preclinical and Clinical Studies Supporting Development of Drug + Digital Combination Therapies for Chronic Diseases

by Zack Biskupiak, Victor Vinh Ha, Aarushi Rohaj and Grzegorz Bulaj

J. Clin. Med. 2024, 13(2), 403; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm13020403 - 11 Jan 2024

Viewed by 3973

Abstract

Limitations of pharmaceutical drugs and biologics for chronic diseases (e.g., medication non-adherence, adverse effects, toxicity, or inadequate efficacy) can be mitigated by mobile medical apps, known as digital therapeutics (DTx). Authorization of adjunct DTx by the US Food and Drug Administration and draft guidelines on “prescription drug use-related software” illustrate opportunities to create drug + digital combination therapies, ultimately leading towards drug–device combination products (DTx has a status of medical devices). Digital interventions (mobile, web-based, virtual reality, and video game applications) demonstrate clinically meaningful benefits for people living with Alzheimer’s disease, dementia, rheumatoid arthritis, cancer, chronic pain, epilepsy, depression, and anxiety. In the respective animal disease models, preclinical studies on environmental enrichment and other non-pharmacological modalities (physical activity, social interactions, learning, and music) as surrogates for DTx “active ingredients” also show improved outcomes. In this narrative review, we discuss how drug + digital combination therapies can impact translational research, drug discovery and development, generic drug repurposing, and gene therapies. Market-driven incentives to create drug–device combination products are illustrated by Humira^® (adalimumab) facing a “patent-cliff” competition with cheaper and more effective biosimilars seamlessly integrated with DTx. In conclusion, pharma and biotech companies, patients, and healthcare professionals will benefit from accelerating integration of digital interventions with pharmacotherapies. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

31 pages, 4052 KiB

Open AccessReview

Dissecting Polygenic Etiology of Ischemic Stroke in the Era of Precision Medicine

by Jiang Li, Vida Abedi and Ramin Zand

J. Clin. Med. 2022, 11(20), 5980; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11205980 - 11 Oct 2022

Cited by 2 | Viewed by 3140

Abstract

Ischemic stroke (IS), the leading cause of death and disability worldwide, is caused by many modifiable and non-modifiable risk factors. This complex disease is also known for its multiple etiologies with moderate heritability. Polygenic risk scores (PRSs), which have been used to establish a common genetic basis for IS, may contribute to IS risk stratification for disease/outcome prediction and personalized management. Statistical modeling and machine learning algorithms have contributed significantly to this field. For instance, multiple algorithms have been successfully applied to PRS construction and integration of genetic and non-genetic features for outcome prediction to aid in risk stratification for personalized management and prevention measures. PRS derived from variants with effect size estimated based on the summary statistics of a specific subtype shows a stronger association with the matched subtype. The disruption of the extracellular matrix and amyloidosis account for the pathogenesis of cerebral small vessel disease (CSVD). Pathway-specific PRS analyses confirm known and identify novel etiologies related to IS. Some of these specific PRSs (e.g., derived from endothelial cell apoptosis pathway) individually contribute to post-IS mortality and, together with clinical risk factors, better predict post-IS mortality. In this review, we summarize the genetic basis of IS, emphasizing the application of methodologies and algorithms used to construct PRSs and integrate genetics into risk models. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

33 pages, 3716 KiB

Open AccessEditor’s ChoiceReview

Clinical Applications of Artificial Intelligence—An Updated Overview

by Ștefan Busnatu, Adelina-Gabriela Niculescu, Alexandra Bolocan, George E. D. Petrescu, Dan Nicolae Păduraru, Iulian Năstasă, Mircea Lupușoru, Marius Geantă, Octavian Andronic, Alexandru Mihai Grumezescu and Henrique Martins

J. Clin. Med. 2022, 11(8), 2265; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm11082265 - 18 Apr 2022

Cited by 47 | Viewed by 6992

Abstract

Artificial intelligence has the potential to revolutionize modern society in all its aspects. Encouraged by the variety and vast amount of data that can be gathered from patients (e.g., medical images, text, and electronic health records), researchers have recently increased their interest in developing AI solutions for clinical care. Moreover, a diverse repertoire of methods can be chosen towards creating performant models for use in medical applications, ranging from disease prediction, diagnosis, and prognosis to opting for the most appropriate treatment for an individual patient. In this respect, the present paper aims to review the advancements reported at the convergence of AI and clinical care. Thus, this work presents AI clinical applications in a comprehensive manner, discussing the recent literature studies classified according to medical specialties. In addition, the challenges and limitations hindering AI integration in the clinical setting are further pointed out. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

37 pages, 6682 KiB

Open AccessReview

Artificial Intelligence: A Shifting Paradigm in Cardio-Cerebrovascular Medicine

by Vida Abedi, Seyed-Mostafa Razavi, Ayesha Khan, Venkatesh Avula, Aparna Tompe, Asma Poursoroush, Alireza Vafaei Sadr, Jiang Li and Ramin Zand

J. Clin. Med. 2021, 10(23), 5710; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm10235710 - 06 Dec 2021

Cited by 6 | Viewed by 4296

Abstract

The future of healthcare is an organic blend of technology, innovation, and human connection. As artificial intelligence (AI) is gradually becoming a go-to technology in healthcare to improve efficiency and outcomes, we must understand our limitations. We should realize that our goal is not only to provide faster and more efficient care, but also to deliver an integrated solution to ensure that the care is fair and not biased to a group of sub-population. In this context, the field of cardio-cerebrovascular diseases, which encompasses a wide range of conditions—from heart failure to stroke—has made some advances to provide assistive tools to care providers. This article aimed to provide an overall thematic review of recent development focusing on various AI applications in cardio-cerebrovascular diseases to identify gaps and potential areas of improvement. If well designed, technological engines have the potential to improve healthcare access and equitability while reducing overall costs, diagnostic errors, and disparity in a system that affects patients and providers and strives for efficiency. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

Other

Jump to: Research, Review

13 pages, 1666 KiB

Open AccessSystematic Review

Many Models, Little Adoption—What Accounts for Low Uptake of Machine Learning Models for Atrial Fibrillation Prediction and Detection?

by Yuki Kawamura, Alireza Vafaei Sadr, Vida Abedi and Ramin Zand

J. Clin. Med. 2024, 13(5), 1313; https://0-doi-org.brum.beds.ac.uk/10.3390/jcm13051313 - 26 Feb 2024

Viewed by 686

Abstract

(1) Background: Atrial fibrillation (AF) is a major risk factor for stroke and is often underdiagnosed, despite being present in 13–26% of ischemic stroke patients. Recently, a significant number of machine learning (ML)-based models have been proposed for AF prediction and detection for primary and secondary stroke prevention. However, clinical translation of these technological innovations to close the AF care gap has been scant. Herein, we sought to systematically examine studies, employing ML models to predict incident AF in a population without prior AF or to detect paroxysmal AF in stroke cohorts to identify key reasons for the lack of translation into the clinical workflow. We conclude with a set of recommendations to improve the clinical translatability of ML-based models for AF. (2) Methods: MEDLINE, Embase, Web of Science, Clinicaltrials.gov, and ICTRP databases were searched for relevant articles from the inception of the databases up to September 2022 to identify peer-reviewed articles in English that used ML methods to predict incident AF or detect AF after stroke and reported adequate performance metrics. The search yielded 2815 articles, of which 16 studies using ML models to predict incident AF and three studies focusing on ML models to detect AF post-stroke were included. (3) Conclusions: This study highlights that (1) many models utilized only a limited subset of variables available from patients’ health records; (2) only 37% of models were externally validated, and stratified analysis was often lacking; (3) 0% of models and 53% of datasets were explicitly made available, limiting reproducibility and transparency; and (4) data pre-processing did not include bias mitigation and sufficient details, leading to potential selection bias. Low generalizability, high false alarm rate, and lack of interpretability were identified as additional factors to be addressed before ML models can be widely deployed in the clinical care setting. Given these limitations, our recommendations to improve the uptake of ML models for better AF outcomes include improving generalizability, reducing potential systemic biases, and investing in external validation studies whilst developing a transparent modeling pipeline to ensure reproducibility. Full article

(This article belongs to the Special Issue The Convergence of Human and Artificial Intelligence on Clinical Care - Part II)

► Show Figures

Figure 1

Journal Menu

Journal Browser

The Convergence of Human and Artificial Intelligence on Clinical Care - Part II

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Published Papers (11 papers)

Research

Review

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI