bims-aukdir Biomed News
on Automated knowledge discovery in diabetes research
Issue of 2025–06–01
seven papers selected by
Mott Given



  1. Med Eng Phys. 2025 Jun;pii: S1350-4533(25)00069-4. [Epub ahead of print]140 104350
      Diabetic retinopathy (DR) is a retinal affliction in patients suffering from diabetes. If DR is unidentified at an earlier stage, it may lead to blindness. Manual screening of DR using fundus images is a complex and time-consuming task. In the past, many automated techniques have been developed for DR detection and classification. In the case of multiclass fundus images, producing reliable classification performance is a challenge for researchers. Hence, this paper presents a novel transfer learning-based approach to classify DR using fundus images. The proposed technique is based on EfficientNetB3 with squeeze and excitation block. EfficientNetB3 performs classification tasks very well using an effective architecture with fewer parameters while the squeeze and excitation block improves the model's ability by focusing on crucial features. For experimentation of the proposed technique, fundus images of the APTOS-2019 dataset are utilized. The proposed technique achieves overall 88.44% accuracy, 98.00% specificity, 84.00% precision, 83.00% sensitivity, 83.00% F1-score, and 0.88 kappa score for all five classes of fundus images of the APTOS-2019 dataset. In addition to this, the proposed technique is also experimented using fundus images of the IDRiD and Messidor-2 datasets. The performance of the proposed technique is better than many existing DR detection techniques.
    Keywords:  Diabetic retinopathy; EfficientNetB3; Fundus images; Squeeze and excitation block
    DOI:  https://doi.org/10.1016/j.medengphy.2025.104350
  2. AMIA Annu Symp Proc. 2024 ;2024 1099-1108
      Integrating Electronic Health Records (EHR) and the application of machine learning present opportunities for enhancing the accuracy and accessibility of data-driven diabetes prediction. In particular, developing data-driven machine learning models can provide early identification of patients with high risk for diabetes, potentially leading to more effective therapeutic strategies and reduced healthcare costs. However, regulation restrictions create barriers to developing centralized predictive models. This paper addresses the challenges by introducing a federated learning approach, which amalgamates predictive models without centralized data storage and processing, thus avoiding privacy issues. This marks the first application of federated learning to predict diabetes using real clinical datasets in Canada extracted from the Canadian Primary Care Sentinel Surveillance Network (CPCSSN) without cross-province patient data sharing. We address class-imbalance issues through downsampling techniques and compare federated learning performance against province-based and centralized models. Experimental results show that the federated MLP model presents a similar or higher performance compared to the model trained with the centralized approach. However, the federated logistic regression model showed inferior performance compared to its centralized peer.
  3. Eye (Lond). 2025 May 30.
       BACKGROUND/OBJECTIVES: Artificial intelligence (AI) assessment of diabetic retinopathy (DR) instead of scarce trained specialists could potentially increases the efficiency and accessibility of screening programs. This systematic review aims to systematically examine the uptake of follow-up appointments with initial computer-based AI and human graders of DR.
    METHODS: We conducted a systematic review and meta-analysis by screening articles in any languages in PubMed, MEDLINE (Ovid), EMBASE, Web of Science, Cochrane CENTRAL and CDSR published from database inception up to 20th August 2024. We used random-effects meta-analysis to pool the results as odds ratios (OR) with corresponding 95% confidence intervals (CI).
    RESULTS: Data from a total of 20,108 patients with diabetes (6476 participants graded using AI and 13,632 participants graded by human-graders; age range of the participants 5 to 67 years) from six studies were included. The result of the pooled meta-analysis showed that initial AI assessment of DR significantly increased uptake of follow-up appointments compared to human grader-based (OR = 1.89, 95% CI 1.78-2.01, P = 0.00001).
    CONCLUSIONS: The present systematic review and meta-analysis suggest that initial AI-based algorithm for screening DR is associated with an increased uptake of follow-up examination. This is most likely due to instant results being made available with AI based algorithms when compared to a delay in assessment with human graders.
    DOI:  https://doi.org/10.1038/s41433-025-03849-4
  4. EPMA J. 2025 Jun;16(2): 519-533
       Objective: Diabetes and hypertension pose significant health risks, especially when poorly managed. Retinal evaluation though fundus photography can provide non-invasive assessment of these diseases, yet prior studies focused on disease presence, overlooking control statuses. This study evaluated vision transformer (ViT)-based models for assessing the presence and control statuses of diabetes and hypertension from retinal images.
    Methods: ViT-based models with ResNet-50 for patch projection were trained on images from the UK Biobank (n = 113,713) and Singapore Epidemiology of Eye Diseases study (n = 17,783), and externally validated on the Singapore Prospective Study Programme (n = 7,793) and the Beijing Eye Study (n = 6064). Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC) for multiple tasks: detecting disease, identifying poorly controlled and well-controlled cases, distinguishing between poorly and well-controlled cases, and detecting pre-diabetes or pre-hypertension.
    Results: The models demonstrated strong performance in detecting disease presence, with AUROC values of 0.820 for diabetes and 0.781 for hypertension in internal testing. External validation showed AUROCs ranging from 0.635 to 0.755 for diabetes, and 0.727 to 0.832 for hypertension. For identifying poorly controlled cases, the performance remained high with AUROCs of 0.871 (internal) and 0.655-0.851 (external) for diabetes, and 0.853 (internal) and 0.792-0.915 (external) for hypertension. Detection of well-controlled cases also yielded promising results for diabetes (0.802 [internal]; 0.675-0.838 [external]), and hypertension (0.740 [internal] and 0.675-0.807 [external]). In distinguishing between poorly and well-controlled disease, AUROCs were more modest with 0.630 (internal) and 0.512-0.547 (external) for diabetes, and 0.651 (internal) and 0.639-0.683 (external) for hypertension. For pre-disease detection, the models achieved AUROCs of 0.746 (internal) and 0.523-0.590 (external) for pre-diabetes, and 0.669 (internal) and 0.645-0.679 (external) for pre-hypertension.
    Conclusion: ViT-based models show promise in classifying the presence and control statuses of diabetes and hypertension from retinal images. These findings support the potential of retinal imaging as a tool in primary care for opportunistic detection of diabetes and hypertension, risk stratification, and individualised treatment planning. Further validation in diverse clinical settings is warranted to confirm practical utility.
    Supplementary Information: The online version contains supplementary material available at 10.1007/s13167-025-00412-9.
    Keywords:  Deep learning; Diabetes; Health risk assessment; Hypertension; Improved individual outcomes; Innovative screening programs; Opportunistic screening; Predictive Preventive Personalized Medicine (PPPM / 3PM); Preventable diseases; Protection against health-to-disease transition; Retinal image; Risk stratification
    DOI:  https://doi.org/10.1007/s13167-025-00412-9
  5. Sci Rep. 2025 May 25. 15(1): 18143
      Hypoglycemia is a serious complication in individuals with type 2 diabetes mellitus. Identifying who is most at risk remains challenging due to the non-linear relationships between hypoglycemia and its associated risk factors. The objective of this study is to evaluate the importance and impact of risk factors related to the incidence of hypoglycemia through an explainable machine learning method. This prospective study enrolled 1306 adults with type 2 diabetes mellitus at a specialized diabetes center. Over three months, participants were asked to do self-monitoring blood glucose measurements and record hypoglycemic events. Nine clinically relevant features were analyzed using five machine learning models. The performance of the models was evaluated by different metrics. The SHapley Additive exPlanation method was used to elucidate how each covariate influenced the risk of hypoglycemia. Overall, 419 participants (32.08%) reported at least one hypoglycemic episode. Our findings highlight the non-linear nature of hypoglycemia risk in individuals with T2DM. Insulin therapy, Diabetes duration (> 13.7 years), and eGFR (< 60.2 mL/min/1.73 m2) were the most important predictors of hypoglycemia, followed by age, HbA1C, triglycerides, total cholesterol, gender, and BMI.
    Keywords:  Hypoglycemia prediction; Machine learning; SHAP; Type 2 diabetes mellitus
    DOI:  https://doi.org/10.1038/s41598-025-03030-7
  6. Indian J Ophthalmol. 2025 Jun 01. 73(6): 797-806
      Recent advances in deep learning and machine learning have greatly increased the capabilities of extracting features for evaluating the response to anti VEGF treatment in patients with Diabetic Macular Edema (DME). In this review, we explore how these algorithms can be used for discriminating between responders and non-responders to anti vascular endothelial growth factor (VEGF) injections. Electronic databases, including PubMed, IEEE Xplore, BioMed, JAMA, and Google Scholar, were searched, and reference lists from relevant publications were also considered from inception till August 31, 2023, based on the inclusion and exclusion criteria. Data extraction was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The results focus on keywords such as DME, OCT, anti VEGF, and patient responses after anti VEGF injections. The article measures the effectiveness of different machine learning and deep learning algorithms, including linear discriminant analysis (LDA), ResNet-50, CNN with attention, quadratic discriminant analysis (QDA), random forest (RF), and support vector machines (SVM), in analyzing eyes that could tolerate extended interval dosing. According to a review of 50 relevant papers published between 2016 and 2023, the algorithms achieved an average automated sensitivity of 74% (95% CI: 0.55-0.92) in detecting treatment responses.
    Keywords:  Anti-VEGF treatment; diabetic retinopathy; patient response metrics; retinal vascular permeability
    DOI:  https://doi.org/10.4103/IJO.IJO_1810_24
  7. Sensors (Basel). 2025 May 20. pii: 3207. [Epub ahead of print]25(10):
      Continuous monitoring of glucose levels is important for diabetes management and prevention. While traditional glucose monitoring methods are often invasive and expensive, recent approaches using machine learning (ML) models have explored non-invasive alternatives-but many still depend on manually logged food intake and activity, which is burdensome and impractical for everyday use. In this study, we propose a novel approach that eliminates the need for manual input by utilizing only passively collected, automatically recorded multi-modal data from non-invasive wearable sensors. This enables practical and continuous glucose prediction in real-world, free-living environments. We used the BIG IDEAs Lab Glycemic Variability and Wearable Device Data (BIGIDEAs) dataset, which includes approximately 26,000 CGM readings, simultaneous ly collected wearable data, and demographic information. A total of 236 features encompassing physiological, behavioral, circadian, and demographic factors were constructed. Feature selection was conducted using random-forest-based importance analysis to select the most relevant features for model training. We evaluated the effectiveness of various ML regression techniques, including linear regression, ridge regression, random forest regression, and XGBoost regression, in terms of prediction and clinical accuracy. Biological sex, circadian rhythm, behavioral features, and tonic features of electrodermal activity (EDA) emerged as key predictors of glucose levels. Tree-based models outperformed linear models in both prediction and clinical accuracy. The XGBoost (XR) model performed best, achieving an R-squared of 0.73, an RMSE of 11.9 mg/dL, an NRMSE of 0.52 mg/dL, a MARD of 7.1%, and 99.4% of predictions falling within Zones A and B of the Clarke Error Grid. This study demonstrates the potential of combining feature engineering and tree-based ML regression techniques for continuous glucose monitoring using wearable sensors.
    Keywords:  continuous glucose monitoring; machine learning; multi-modal; non-invasive; wearable sensors
    DOI:  https://doi.org/10.3390/s25103207