Journal of Diabetes Science and Technology最新文献

{"title":"Hypoglycemia Prediction in Type 1 Diabetes With Electrocardiography Beat Ensembles.","authors":"Mu-Ruei Tseng, Kathan Vyas, Anurag Das, Waris Quamer, Darpit Dave, Madhav Erranguntla, Carolina Villegas, Daniel DeSalvo, Siripoom McKay, Gerard Cote, Ricardo Gutierrez-Osuna","doi":"10.1177/19322968251319347","DOIUrl":"10.1177/19322968251319347","url":null,"abstract":"<p><strong>Introduction: </strong>Current methods to detect hypoglycemia in type 1 diabetes (T1D) require invasive sensors (ie, continuous glucose monitors, CGMs) that generally have low accuracy in the hypoglycemic range. A forward-looking alternative is to monitor physiological changes induced by hypoglycemia that can be measured non-invasively using, eg, electrocardiography (ECG). However, current methods require extraction of fiduciary points in the ECG signal (eg, to estimate QT interval), which is challenging in ambulatory settings.</p><p><strong>Methods: </strong>To address this issue, we present a machine-learning model that uses (1) convolutional neural networks (CNNs) to extract morphological information from raw ECG signals without the need to identify fiduciary points and (2) ensemble learning to aggregate predictions from multiple ECG beats. We evaluate the model on an experimental data set that contains ECG and CGM recordings over a period of 14 days from ten participants with T1D. We consider two testing scenarios, one that divides ECG data according to CGM readings (CGM-split) and another that divides ECG data on a day-to-day basis (day-split).</p><p><strong>Results: </strong>We find that models trained using CGM-splits tend to produce overly optimistic estimates of hypoglycemia prediction, whereas day-splits provide more realistic estimates, which are consistent with the intrinsic accuracy of CGM devices. More importantly, we find that aggregating predictions from multiple ECG beats using ensemble learning significantly improves predictions at the beat level, though these improvements have large inter-individual differences.</p><p><strong>Conclusion: </strong>Deep learning models and ensemble learning can extract and aggregate morphological information in ECG signals that is predictive of hypoglycemia. Using two validation procedures, we estimate an upper bound on the accuracy of ECG hypoglycemia prediction of 81% equal error rate and a lower bound of 60%. Further improvements may be achieved using big-data approaches that require longitudinal data from a large cohort of participants.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251319347"},"PeriodicalIF":4.1,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11863193/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143501466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Toward Detection of Nocturnal Hypoglycemia in People With Diabetes Using Consumer-Grade Smartwatches and a Machine Learning Approach.","authors":"Camilo Mendez, Ceren Asli Kaykayoglu, Thiemo Bähler, Juri Künzler, Aritz Lizoain, Martina Rothenbühler, Markus H Schmidt, Markus Laimer, Lilian Witthauer","doi":"10.1177/19322968251319800","DOIUrl":"10.1177/19322968251319800","url":null,"abstract":"<p><strong>Background: </strong>Nocturnal hypoglycemia poses significant risks to individuals with insulin-treated diabetes, impacting health and quality of life. Although continuous glucose monitoring (CGM) systems reduce these risks, their poor accuracy at low glucose levels, high cost, and availability limit their use. This study examined physiological biomarkers associated with nocturnal hypoglycemia and evaluated the use of machine learning (ML) to detect hypoglycemia during nighttime sleep using data from consumer-grade smartwatches.</p><p><strong>Methods: </strong>This study analyzed 351 nights of 36 adults with insulin-treated diabetes. Participants wore two smartwatches alongside CGM systems. Linear mixed-effects models compared sleep and vital signs between nights with and without hypoglycemia during early and late sleep. A ML model was trained to detect hypoglycemia solely using smartwatch data.</p><p><strong>Results: </strong>Sixty-six nights with spontaneous hypoglycemia were recorded. Hypoglycemic nights showed increased wake periods, heart rate, stress levels, and activity during early sleep, with weaker effects during late sleep. In nights when hypoglycemia occurred during early sleep, the ML model performed comparable or better than prior studies with an area under the receiver operator curve of 0.78 for level 1 and 0.83 for level 2 hypoglycemia, with sensitivity of 0.78 and 0.89, specificity of 0.64 for both, negative predictive value of 0.94 and 0.99, and positive predictive value of 0.25 and 0.13 for level 1 and level 2 hypoglycemia, respectively.</p><p><strong>Conclusions: </strong>Consumer-grade smartwatches demonstrate promise for detecting nocturnal hypoglycemia, particularly during early sleep. Refining models to reduce false alarms could enhance their clinical utility as low-cost, accessible tools to complement CGM.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251319800"},"PeriodicalIF":4.1,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11851596/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143492230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Validation of the Diabetes Technology Society Error Grid.","authors":"Jan S Krouwer","doi":"10.1177/19322968251320653","DOIUrl":"10.1177/19322968251320653","url":null,"abstract":"","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251320653"},"PeriodicalIF":4.1,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11851577/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143492231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feasibility of a Glucose Manipulation Procedure for the Standardized Performance Evaluation of Continuous Glucose Monitoring Systems.","authors":"Manuela Link, Manuel Eichenlaub, Delia Waldenmaier, Stephanie Wehrstedt, Stefan Pleus, Nina Jendrike, Sükrü Öter, Cornelia Haug, Stefanie Hossmann, Martina Rothenbühler, Derek Brandt, Guido Freckmann","doi":"10.1177/19322968251317526","DOIUrl":"10.1177/19322968251317526","url":null,"abstract":"<p><strong>Background: </strong>In continuous glucose monitoring (CGM) system performance studies, it is common to implement specific procedures for manipulating the participants' blood glucose (BG) levels during the collection of comparator BG measurements. Recently, such a procedure was proposed by a group of experts, and this study assessed its ability to produce combinations of BG levels and rates of change (RoCs) with certain characteristics.</p><p><strong>Methods: </strong>During three separate in-clinic sessions conducted over 15 days, capillary BG measurements were carried out every 15 minutes for 7 hours. Simultaneously, the participants' BG levels were manipulated by controlling food intake and insulin administration to induce transient hyperglycemia and hypoglycemia. Subsequently, the combinations of BG levels and RoCs were categorized into dynamic glucose regions distinguishing between rapidly increasing BG levels (Alert high), hyperglycemia (BG high), rapidly falling BG levels (Alert low), and hypoglycemia (BG low).</p><p><strong>Results: </strong>A total of 24 adult participants with type 1 diabetes were included. Capillary BG-RoC combinations showed 7.5% in the Alert high region, 13.3% in the BG high region, 9.8% in the Alert low region, and 11.0% in the BG low region. No adverse events related to the glucose manipulation procedure were documented.</p><p><strong>Conclusions: </strong>As recommended by the experts, the percentage of data points in regions was ≥7.5%, demonstrating the procedure's feasibility. However, given that the recommendation for the alert high region was only barely achieved, we suggest optimizations to the procedure and definition of dynamic glucose regions to facilitate the procedures' adoption in standardized CGM performance evaluations.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251317526"},"PeriodicalIF":4.1,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11848859/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143483327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of the Effectiveness of Continuous Glucose Monitors on Glycemic Control in Patients With Type 2 Diabetes Receiving Institutional Financial Assistance.","authors":"Kevin K Cheng, Maxine F Vera Cruz, Tracy S Tylee, Mary S Kelly","doi":"10.1177/19322968251320122","DOIUrl":"10.1177/19322968251320122","url":null,"abstract":"<p><strong>Background: </strong>Current guidelines suggest utilizing continuous glucose monitoring (CGM) to improve hemoglobin A_1c (HbA_1c) in patients with diabetes. Financial cost remains a barrier to implementation. Medicare coverage criteria include all patients with diabetes treated with at least one injection of insulin per day, while Washington Medicaid is more restrictive. There remains a paucity of literature examining effectiveness of CGMs on clinical outcomes among patients with type 2 diabetes with lower incomes.</p><p><strong>Methods: </strong>This is a single-center, retrospective, observational study including adults with type 2 diabetes receiving institutional financial assistance for CGMs. A cohort with no CGM use is included for comparison. The primary outcome is change in HbA_1c approximately three months after CGM implementation from baseline. Secondary outcomes include mean differences in number of antidiabetic agents and changes in insulin dose prior to and after CGM implementation.</p><p><strong>Results: </strong>Among the CGM cohort, most patients were of Hispanic ethnicity (77%) and a majority had no insurance (77%). The average HbA_1c prior to CGM implementation was 8.3% and three months post-CGM was 7.7%, with a mean difference of -0.6% (<i>P</i> = .004). There were no statistically significant differences in the average number of antidiabetic agents, total daily dosages of insulin, or mean differences in the number of emergency room visits or hospitalizations prior to and post-implementation of a CGM.</p><p><strong>Conclusion: </strong>Overall, there is a statistical and clinical improvement in HbA_1c before and after implementation of CGMs in patients with type 2 diabetes who meet Medicaid criteria for CGM coverage receiving financial assistance.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251320122"},"PeriodicalIF":4.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11843561/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143468223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Processing Algorithm to Address Real-World Data Quality Issues With Continuous Glucose Monitoring Data.","authors":"Walter Williamson, Joyce M Lee, Irina Gaynanova","doi":"10.1177/19322968251319801","DOIUrl":"10.1177/19322968251319801","url":null,"abstract":"<p><p>Continuous glucose monitoring (CGM) data stored in data warehouses often include duplicated or time-shifted uploads from the same patient, compromising data quality and accuracy of resulting CGM metrics. We developed a processing algorithm to detect and resolve these errors. We validated the algorithm using two weeks of CGM data from 2038 patients with diabetes. Duplication errors were identified in 528 patients, with 25.7% showing significant differences in at least one metric (Time in Range, Coefficient of Variation, Glycemic Management Indicator, or Glycemic Episode counts) between raw and processed data. Eleven patients crossed clinically meaningful thresholds in one or more metrics after processing. Our results underscore the importance of real-world CGM data processing to maintain accurate and reliable CGM metrics for research and clinical care.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251319801"},"PeriodicalIF":4.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11843558/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143468217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Plantar Thermogram Analysis Using Deep Learning for Diabetic Foot Risk Classification.","authors":"Vipawee Panamonta, Ratanaporn Jerawatana, Prapai Ariyaprayoon, Panu Looareesuwan, Benyapa Ongphiphadhanakul, Chutintorn Sriphrapradang, Laor Chailurkit, Boonsong Ongphiphadhanakul","doi":"10.1177/19322968251316563","DOIUrl":"10.1177/19322968251316563","url":null,"abstract":"<p><strong>Aims: </strong>Thermography is a noninvasive method to identify patients at risk of diabetic foot ulcers. In this study, we employed thermography and deep learning to stratify patients with diabetes at risk of developing foot ulcers.</p><p><strong>Methods: </strong>We prospectively recorded clinical data and plantar thermograms for adult patients with diabetes who underwent diabetic foot screening. A total of 153 thermal images were analyzed using a deep learning algorithm to determine the risk of diabetic foot ulcers. The neural network was trained using a balanced dataset consisting of 98 thermal images (49 normal and 49 abnormal), with 80% allocated for training and 20% for validation. The trained model was then validated on a separate testing dataset consisting of 55 thermal images (42 normal and 13 abnormal). The neural network was trained to prioritize higher sensitivity in identifying at-risk feet for screening purposes.</p><p><strong>Results: </strong>Participants had a mean age of 63.1 ± 12.6 years (52.3% female), and 62.1% had been diagnosed with diabetes for more than 10 years. The average body mass index was 27.5 ± 5.6 kg/m². Of the thermal images, 91 were classified as category 0 and 62 as categories 1 to 3, according to the diabetic foot risk classification system of the International Working Group on the Diabetic Foot. Using five-fold cross-validation, the neural network model achieved an overall accuracy of 71.8 ± 4.9%, a sensitivity of 81.2 ± 10.0%, and a specificity of 64.0 ± 7.4%. Additionally, the Matthews correlation coefficient was 0.46 ± 0.08.</p><p><strong>Conclusions: </strong>These results suggest that thermography combined with deep learning could be developed for screening purposes to stratify patients at risk of developing diabetic foot ulcers.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251316563"},"PeriodicalIF":4.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11843557/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143468225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of the Effect of Teleconsultations, Hybrid Visits, and In-Person Visits on Glycemic and Metabolic Parameters Among Individuals With Type 2 Diabetes in India.","authors":"Anandakumar Amutha, Shyama Reji, Ramamurthy Hema Aarthi, Srinivas Keertan Rao, S Ganesan, Saravanan Jebarani, Gangadhara Praveen, Ranjit Unnikrishnan, Viswanathan Mohan, Ranjit Mohan Anjana","doi":"10.1177/19322968251319333","DOIUrl":"10.1177/19322968251319333","url":null,"abstract":"<p><strong>Aim: </strong>We compared biochemical and clinical data of individuals with type 2 diabetes (T2D) who opted for only teleconsultation (ie, no in-person visit at all), hybrid visits (combining home blood tests and in-person consultation), and fully in-person visits (both tests and consultation in person) at a tertiary care diabetes center.</p><p><strong>Methods: </strong>In this observational cohort study, we retrieved demographic, anthropometric, and biochemical data of 8197 individuals with T2D who sought diabetes care between 2021 and 2023 (384 participants with only teleconsultations, 721 with hybrid visits, and 7092 with fully in-person visits) from the electronic medical records of a chain of tertiary diabetes care centers across India.</p><p><strong>Results: </strong>Individuals who opted for teleconsultation had a shorter duration of diabetes compared with those who opted for hybrid or fully in-person visits. Although participants who opted for a teleconsultation had better glycemic and lipid control at baseline, those who underwent hybrid and in-person visits showed greater improvements in fasting plasma glucose, glycated hemoglobin (A1c), and LDL cholesterol (LDL-C) during follow-up. Improvements in overall ABC target achievement (<u>A</u>1c, <u>B</u>lood pressure, and <u>L</u>DL-C) were greater in participants who had in-person visits compared with the other two groups.</p><p><strong>Conclusion: </strong>While teleconsultation is a useful complement to in-person visits, the latter results in better glycemic and lipid control, perhaps due to more effective engagement with the diabetes care team.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251319333"},"PeriodicalIF":4.1,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11840818/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143449197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Impact of Virtual Consultations on Quality of Care for Patients With Type 2 Diabetes: A Systematic Review and Meta-Analysis.","authors":"Reham Aldakhil, Geva Greenfield, Elena Lammila-Escalera, Liliana Laranjo, Benedict W J Hayhoe, Azeem Majeed, Ana Luísa Neves","doi":"10.1177/19322968251316585","DOIUrl":"10.1177/19322968251316585","url":null,"abstract":"<p><strong>Background: </strong>Virtual consultations (VC) have transformed healthcare delivery, offering a convenient and effective way to manage chronic conditions such as Type 2 Diabetes (T2D). This systematic review and meta-analysis evaluated the impact of VC on the quality of care provided to patients with T2D, mapping it across the six domains of the US National Academy of Medicine (NAM) quality-of-care framework (ie, effectiveness, efficiency, patient-centeredness, timeliness, safety, and equity).</p><p><strong>Methods: </strong>A systematic search was conducted in PubMed/MEDLINE, Cochrane, Embase, CINAHL, and Web of Science for the period between January 2010 and December 2024. Eligible studies involved adult T2D patients, evaluated synchronous VCs, and reported outcomes relevant to NAM quality domains. Two independent reviewers performed screening, and studies were assessed using the Mixed Methods Appraisal Tool (MMAT). A narrative synthesis was conducted for each quality domain, and a meta-analysis of HbA1c levels was performed using random-effects models.</p><p><strong>Results: </strong>In total, 15 studies involving 821 014 participants were included. VCs were comparable with face-to-face care in effectiveness, efficiency, patient-centeredness, and timeliness, with improvements in accessibility and patient satisfaction. Mixed results were found for safety due to limitations in physical assessments, and for equity, with older adults and those with lower digital literacy facing more challenges. The meta-analysis showed no significant difference in HbA1c reduction between VCs and face-to-face (standardized mean difference [SMD] = -0.31, 95% confidence interval [CI]: -0.71 to 0.09, <i>P</i> = 0.12).</p><p><strong>Conclusion: </strong>VCs offer a promising alternative to in-person care, but addressing digital disparities and improving access for older adults are essential for maximizing VC potential.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251316585"},"PeriodicalIF":4.1,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11833803/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143441092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence for Diabetic Foot Screening Based on Digital Image Analysis: A Systematic Review.","authors":"Ni Kadek Indah Sunar Anggreni, Heri Kristianto, Dian Handayani, Yuyun Yueniwati, Paulus Lucky Tirma Irawan, Rulli Rosandi, Rinik Eko Kapti, Avief Destian Purnama","doi":"10.1177/19322968251317521","DOIUrl":"10.1177/19322968251317521","url":null,"abstract":"<p><strong>Introduction: </strong>Early detection of diabetic foot complications is essential for effective management and prevention of complications. Artificial intelligence (AI) technology based on digital image analysis offers a promising noninvasive method for diabetic foot screening. This systematic review aims to identify a study on the development of an AI model for diabetic foot screening using digital image analysis.</p><p><strong>Methodology: </strong>The review scrutinized articles published between 2018 and 2023, sourced from PubMed, ProQuest, and ScienceDirect. The keyword-based search resulted in 2214 relevant articles and nine articles that met the inclusion criteria. The article quality assessment was done through Quality Assessment of Diagnostic Accuracy Studies (QUADAS). Data were extracted and analyzed using NVivo.</p><p><strong>Results: </strong>Thermal imagery or foot thermogram was the main data source, with plantar temperature distribution patterns as an important indicator. Deep learning methods, specifically artificial neural networks (ANNs) and convolutional neural networks (CNNs), are the most commonly used methods. The highest performance is demonstrated by the ANN model with MATLAB's Image Processing Toolbox that is able to classify each type of macula with 97.5% accuracy. The findings show the great potential of AI in improving the accuracy and efficiency of diabetic foot screening.</p><p><strong>Conclusion: </strong>This research provides important insights into the development of AI in digital image-based diabetic foot screening. Future studies need to focus on evaluating clinical applicability, including ethical aspects and patient data security, as well as developing more comprehensive data sets.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251317521"},"PeriodicalIF":4.1,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11833800/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143441082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}