bims-aukdir 2025-07-20 papers

bims-aukdir

Biomed News

on Automated knowledge discovery in diabetes research

Issue of 2025–07–20
fifteen papers selected by
Mott Given

Prediction of diabetic retinopathy using machine learning and its association with dementia risk in older adults with type 2 diabetes mellitus.
Comparison of Validity and Reliability of Manual Consensus Grading vs. Automated AI Grading for Diabetic Retinopathy Screening in Oslo, Norway: A Cross-Sectional Pilot Study.
A Simplified Machine Learning Model for Predicting Reduced Kidney Function in Thai Patients with Type 2 Diabetes: A Retrospective Study.
Enhancing pathological feature discrimination in diabetic retinopathy multi-classification with self-paced progressive multi-scale training.
Comparative analysis of artificial intelligence tools in generating diabetic retinopathy guidelines.
Generation of multidisease fundus photographs with code-free platform.
Performance of a Retinal Imaging Camera With On-Device Intelligence for Primary Care: Retrospective Study.
A participatory approach to deploy responsible artificial intelligence for diabetes prediction and prevention.
Improving T2D machine learning-based prediction accuracy with SNPs and younger age.
Diabetic Tibial Neuropathy Prediction: Improving interpretability of Various Machine-Learning Models Based on Multimodal-Ultrasound Features Using SHAP Methodology.
Prediction of glycaemic control and quality of life in people with type 2 diabetes using glucose-lowering drugs with machine learning-The Maastricht study.
Occupational and Physical Activity Factors as Predictors of Prediabetes: A Machine Learning Study with SHAP-Based Interpretation.
DNA sequence classification for diabetes mellitus using NuSVC and XGBoost: A comparative.
Advancement of artificial intelligence based treatment strategy in type 2 diabetes: A critical update.
Artificial intelligence-based diabetes risk prediction from longitudinal DXA bone measurements.

Diabetes Res Clin Pract. 2025 Jul 15. pii: S0168-8227(25)00392-4. [Epub ahead of print] 112378

Prediction of diabetic retinopathy using machine learning and its association with dementia risk in older adults with type 2 diabetes mellitus.

So Hee Lee, Gyubeom Hwang, Dong Yun Lee, Ja Young Jeon, Seung-Jin Kwag, Seo Young Sohn, Sang Joon Park, Dughyun Choi, Sang Youl Rhee, Rae Woong Park.

   AIMS: Diabetic Retinopathy (DR), a common microvascular complication of diabetes, has been associated with an increased risk of dementia. This study aimed to develop Machine Learning (ML) models to predict DR occurrence and evaluate its potential as a prognostic biomarker for dementia.
METHODS: We included 27,929 patients aged ≥ 50 years newly diagnosed with type 2 diabetes mellitus without prior dementia or eye disease. Prediction models for DR within one year were developed using three ML algorithms: extreme gradient boosting (XGBoost), random forest, and least absolute shrinkage and selection operator. The best-performing model was externally validated across eight institutions. Patients were followed for three years to assess dementia incidence. Dementia risk between ML-predicted DR and non-DR groups was compared using Kaplan-Meier and Cox regression, with results pooled via meta-analysis.
RESULTS: XGBoost demonstrated the best performance (AUROC: 0.746), with external validation AUROCs ranging from 0.555 to 0.620. Predicted DR was significantly associated with increased all-cause dementia risk (HR: 1.32, 95% confidence interval [CI] 1.12-1.56), Alzheimer's disease (HR: 1.30, 95% CI 1.07-1.58), and vascular dementia (HR: 1.38, 95% CI 1.12-1.69).
CONCLUSIONS: ML-predicted DR was significantly associated with future dementia, highlighting its value in early risk stratification among patients with diabetes.

Keywords:  Dementia; Diabetes mellitus; Diabetic retinopathy; Predictive model

DOI:  https://doi.org/10.1016/j.diabres.2025.112378
J Clin Med. 2025 Jul 07. pii: 4810. [Epub ahead of print]14(13):

Comparison of Validity and Reliability of Manual Consensus Grading vs. Automated AI Grading for Diabetic Retinopathy Screening in Oslo, Norway: A Cross-Sectional Pilot Study.

Mia Karabeg, Goran Petrovski, Katrine Holen, Ellen Steffensen Sauesund, Dag Sigurd Fosmark, Greg Russell, Maja Gran Erke, Vallo Volke, Vidas Raudonis, Rasa Verkauskiene, Jelizaveta Sokolovska, Morten Carstens Moe, Inga-Britt Kjellevold Haugen, Beata Eva Petrovski.

  Background: Diabetic retinopathy (DR) is a leading cause of visual impairment worldwide. Manual grading of fundus images is the gold standard in DR screening, although it is time-consuming. Artificial intelligence (AI)-based algorithms offer a faster alternative, though concerns remain about their diagnostic reliability. Methods: A cross-sectional pilot study among patients (≥18 years) with diabetes was established for DR and diabetic macular edema (DME) screening at the Oslo University Hospital (OUH), Department of Ophthalmology, and the Norwegian Association of the Blind and Partially Sighted (NABP). The aim of the study was to evaluate the validity (accuracy, sensitivity, specificity) and reliability (inter-rater agreement) of automated AI-based compared to manual consensus (MC) grading of DR and DME, performed by a multidisciplinary team of healthcare professionals. Grading of DR and DME was performed manually and by EyeArt (Eyenuk) software version v2.1.0, based on the International Clinical Disease Severity Scale (ICDR) for DR. Agreement was measured by Quadratic Weighted Kappa (QWK) and Cohen's Kappa (κ). Sensitivity, specificity, and diagnostic test accuracy (Area Under the Curve (AUC)) were also calculated. Results: A total of 128 individuals (247 eyes) (51 women, 77 men) were included, with a median age of 52.5 years. Prevalence of any vs. referable DR (RDR) was 20.2% vs. 11.7%, while sensitivity was 94.0% vs. 89.7%, specificity was 72.6% was 83.0%, and AUC was 83.5% vs. 86.3%, respectively. DME was detected only in one eye by both methods. Conclusions: AI-based grading offered high sensitivity and acceptable specificity for detecting DR, showing moderate agreement with manual assessments. Such grading may serve as an effective screening tool to support clinical evaluation, while ongoing training of human graders remains essential to ensure high-quality reference standards for accurate diagnostic accuracy and the development of AI algorithms.

Keywords:  EyeArt; artificial intelligence (AI); automated grading; diabetic macular edema; diabetic retinopathy; diagnostic accuracy; fundus photography; manual consensus grading; screening program

DOI:  https://doi.org/10.3390/jcm14134810
J Clin Med. 2025 Jul 04. pii: 4735. [Epub ahead of print]14(13):

A Simplified Machine Learning Model for Predicting Reduced Kidney Function in Thai Patients with Type 2 Diabetes: A Retrospective Study.

Wanjak Pongsittisak, Swangjit Suraamornkul.

  Background: Chronic kidney disease (CKD) is a prevalent complication among individuals with type 2 diabetes (T2D), posing significant diagnostic challenges in resource-limited settings due to infrequent testing and missed hospital visits. This study aimed to develop a simple, effective ML model to identify T2D patients at high risk for reduced kidney function. Methods: We retrospectively analyzed data from 3471 T2D patients collected over a ten-year period at a university hospital in Bangkok, Thailand. Two models were developed using readily available clinical features: one including hemoglobin A1c (HbA1c) levels (the "with-HbA1c" model) and one excluding HbA1c levels (the "non-HbA1c" model). Three tree-based ML algorithms-decision tree, random forest, and extreme gradient boosting (XGBoost) algorithms-were employed. The outcome label was CKD, defined as an estimated Glomerular Filtration Rate (eGFR) < 60 mL/min/1.73 m2 that persisted for more than 90 days. The model performance was evaluated using the AUROC. The feature importance was assessed using Shapley additive explanations (SHAP). Results: The XGBoost algorithm demonstrated a strong predictive performance. The "with-HbA1c" model achieved an AUROC of 0.824, while the "non-HbA1c" model attained a comparable AUROC of 0.819. Both models were well-calibrated. SHAP analysis identified age, HbA1c, and systolic blood pressure as the most influential predictors. Conclusions: Our simplified, interpretable ML models can effectively stratify the risk of reduced kidney function in patients with T2D using minimal, routine data. These models represent a promising step toward integration into clinical practice, such as through EHR-based alerts or patient-facing mobile applications, to improve early CKD detection, particularly in resource-limited settings.

Keywords:  Thailand; chronic kidney disease; machine learning; risk prediction; type 2 diabetes

DOI:  https://doi.org/10.3390/jcm14134735
Sci Rep. 2025 Jul 16. 15(1): 25705

Enhancing pathological feature discrimination in diabetic retinopathy multi-classification with self-paced progressive multi-scale training.

Qiuji Zhou, Yongde Guo, Wenjian Liu, Yifeng Liu, Yanzhen Lin.

Diabetic retinopathy (DR) is a common diabetes complication that presents significant diagnostic challenges due to its reliance on expert assessment and the subtlety of small lesions. Although deep learning has shown promise, its effectiveness is often limited by low-quality data and small sample sizes. To address these issues, we propose a novel deep learning framework for DR that incorporates self-paced progressive learning, introducing training samples from simple to complex, and randomized multi-scale image reconstruction for enhanced data augmentation and feature extraction. Additionally, ensemble learning with Kullback-Leibler (KL) divergence-based collaborative regularization improves classification consistency. The method's effectiveness is demonstrated through experiments on the integrated APTOS and MESSIDOR-Kaggle dataset, achieving an AUC of 0.9907 in 4-class classification, marking a 2.2% improvement compared to the ResNet-50 baseline. Notably, the framework achieves a recall of 97.65% and precision of 96.54% for the No-DR class, and a recall of 98.55% for the Severe class, with precision exceeding 91% across all categories. Furthermore, superior classification performance on limited data samples, as well as robust localization of subtle lesions via multi-scale progressive learning, has been demonstrated, underscoring the potential of the proposed framework for practical clinical deployment.

DOI: https://doi.org/10.1038/s41598-025-07050-1
Adv Ophthalmol Pract Res. 2025 Aug-Sep;5(3):5(3): 180-181

Comparative analysis of artificial intelligence tools in generating diabetic retinopathy guidelines.

Fan Cao, Shengping Hou.

DOI: https://doi.org/10.1016/j.aopr.2025.04.004
Br J Ophthalmol. 2025 Jul 15. pii: bjo-2024-326741. [Epub ahead of print]

Generation of multidisease fundus photographs with code-free platform.

Huiyu Liang, Qi Zhang, Tian Lin, Chenli Hu, Chaoxin Zheng, Xue Yao, Man Chen, Yifan Chen, Yih Chung Tham, Haoyu Chen.

   PURPOSE: To generate fundus photographs of multiple kinds of retinal disease, bypassing the requirement of coding technique.
METHODS: The dataset contained fundus photographs of 10 categories of retinal conditions, with 500 fundus photographs in each category. We randomly divided the collected data into a training set (80%) and a test set (20%). Google Colaboratory was used to implement Pix2Pix to generate fundus photographs for each category. We compared the diagnostic abilities of ophthalmologists on both real and synthetic images. The diagnostic performance of the classification models trained on real, synthetic and combined data sets was also compared. Furthermore, the real and synthesised images were distinguished by ophthalmologists and artificial intelligence (AI) image detection websites.
RESULTS: Fundus photographs of 10 categories were successfully synthesised using our method. The synthetic images showed slightly higher diagnostic accuracy by the three ophthalmologists than the real images (99.7% vs 98.7%, 98.0% vs 96.0% and 99.7% vs 94.3%; p=0.109). Training ResNet-50 and VGG-19 models with a combination of real and synthetic images resulted in significant improvements in accuracy, achieving 93.7% and 89.3%, respectively. Five residents achieved at least 92.5% accuracy in discriminating between real and synthetic images. In contrast, three AI image detection websites showed limited capability in this task, with a maximum accuracy of 51.2%.
CONCLUSION: Pix2Pix on Google Colaboratory can efficiently produce a diverse range of fundus images with typical characters.

Keywords:  Imaging; Retina

DOI:  https://doi.org/10.1136/bjo-2024-326741
JMIR Form Res. 2025 Jul 17. 9 e70331

Performance of a Retinal Imaging Camera With On-Device Intelligence for Primary Care: Retrospective Study.

Matthew Silvestrini, Clarissa Lui, Anil Patwardhan, Ying Chen, Tayyeba Ali, Elie Glik, Honglei Wu, Brian Levinstein, Adrianna Wenz, Nathan Shemonski, Lin Yang, Ian Atkinson, Sam Kavusi.

   Background: Access to screening continues to be a barrier for the early detection of diabetic retinopathy (DR). Primary care-based diabetic retinopathy screening could improve access, but operational challenges, such as cost and workflow management, hamper the widespread adoption of retinal camera systems in primary care clinics in the United States.
Objective: This study aimed to develop and evaluate a retinal screening system suitable for integration into a primary care workflow.
Methods: We developed a nonmydriatic, 45° field imaging retinal camera system, the Verily Numetric Retinal Camera (VNRC; Verily Life Sciences LLC), able to generate high-fidelity retinal images enabled by on-device intelligent features. The VNRC output flows into cloud-based software that accepts and routes digitized images for grading. We evaluated the performance and usability of the VNRC in 2 studies. A retrospective performance study compared the performance of VNRC against a reference camera (Crystalvue NFC-700 [Crystalvue Medical]) as well as the correlation between VNRC capture status and gradability (as determined by ophthalmologist graders). The usability study simulated a primary care setting for a combined cohort of trained and untrained users (corresponding to patients in the simulation) and operators (corresponding to health care personnel in the simulation), where respondents completed a questionnaire about their user experience after attempting to capture images with the VNRC.
Results: In the comparative performance study (N=108, K=206 images), a total of 98.5% (203/206) of images captured by the VNRC were graded as sufficient for clinical interpretation compared to 97.1% (200/206) of Crystalvue NFC-700 images (difference in proportion was 0.015, 95% CI -0.007 to 0.033). In the quality control algorithm evaluation (N=172, K=343 images), we found a positive association (φ=0.58, 95% CI 0.45-0.69) between gradability status (gradable or nongradable) as determined by ophthalmologists and the capture status (recapture not-needed or needed) as determined by the VNRC quality control algorithm. In the usability study (n=15 users and n=30 operators), all participating users (15/15) indicated that they were able to have both eyes screened easily. Most users and operators indicated agreement (from somewhat agree to strongly agree) with statements describing the imaging process as intuitive (15/15, 100% and 29/30, 97%), comfortable (15/15, 100% and 30/30, 100%), and allowing for a positive experience (15/15, 100% and 30/30, 100%), of users and operators, respectively.
Conclusions: Our findings about the performance and usability of this retinal camera system support its deployment as an integrated end-to-end retinal service for primary care. These results warrant additional studies to fully characterize real-world usability across a wider and diverse set of primary care clinics.

Keywords:  diabetic retinopathy; diabetic retinopathy screening; fundus imaging; retinal camera; retinal imaging

DOI:  https://doi.org/10.2196/70331
Digit Health. 2025 Jan-Dec;11:11 20552076251358541

A participatory approach to deploy responsible artificial intelligence for diabetes prediction and prevention.

Laura C Rosella, James Shaw, Shion Guha, Ibukun-Oluwa Omolade Abejirinde, Jennifer L Gibson, Lorraine Lipscombe, Kathy Kornas, Remziye Zaim, Victoria Chui, Ijeoma Uchenna Itanyi.

  Artificial intelligence (AI) technologies have the potential to improve healthcare and public health. Although there has been success in AI for research uses, little progress has been made in implementing health-related AI technologies in health systems. Responsible AI for health systems requires engagement and co-design with health system partners, policymakers, and the community. Deploying responsible AI requires engaging stakeholders, particularly those affected by the technology. This commentary presents the importance of participatory approaches for responsible AI implementation. In this commentary, we discuss the planned use of participatory approaches to responsibly deploying validated machine learning models, with a specific case example of diabetes prediction models that can address the challenge of preventing and managing diabetes in a health system.. The participatory methods engage policy-, provider-, and community-level actors to deploy and implement the AI diabetes tools, inform how AI is implemented in health settings, and overcome common deployment barriers. The future of AI in health settings rests on fine-tuning these practices to enable trust, acceptability, and oversight of these technologies to be deeply established in health systems.

Keywords:  Artificial intelligence; deployment; diabetes; disease; machine learning; participatory

DOI:  https://doi.org/10.1177/20552076251358541
Comput Struct Biotechnol J. 2025 ;27 2772-2781

Improving T2D machine learning-based prediction accuracy with SNPs and younger age.

Cynthia Al Hageh, Andreas Henschel, Hao Zhou, Jorge Zubelli, Moni Nader, Stephanie Chacar, Nantia Iakovidou, Haralampos Hatzikirou, Antoine Abchee, Siobhán O'Sullivan, Pierre A Zalloua.

   Background: This study aimed to evaluate whether integrating clinical and genomic data improves the performance of machine learning (ML) models for predicting Type 2 Diabetes (T2D) risk.
Methods: Six models-Random Forest, Support Vector Machine, Linear Discriminant Analysis, Logistic Regression, Gradient Boosting Machine, and Decision Tree-were trained and tested on a discovery dataset (N=3,546) and validated in the UK Biobank (N=31,620). Model performance was assessed using clinical data alone, combined clinical and genomic data, and in age-specific groups (>55 and ≤55 years).
Results: The inclusion of genomic data modestly improved model performance across all algorithms in the discovery dataset. Clinical features such as family history of T2D and hypertension consistently ranked as top features. When SNPs were added, T2D-associated variants, including rs2943641 (IRS1), rs7903146 (TCF7L2), and rs7756992 (CDKAL1), emerged among the most important features, particularly in younger individuals. These findings demonstrate the translational potential of incorporating genomics for early risk identification. In the UK Biobank, all models achieved AUCs exceeding 91 % with combined clinical and genomic data. Performance was notably better among younger individuals (≤55 years), emphasizing the models' potential for early detection. Integration of a polygenic risk score (PRS) further supported risk prediction, particularly in younger individuals, though incremental gains were modest.
Conclusions: While traditional clinical factors remained the strongest predictors of T2D risk, integration of genomic data produced a modest improvement in model performance, especially among younger adults. Validation across independent datasets confirmed the generalizability of these findings, underscoring the value of multi-dimensional risk-prediction models to refine T2D risk assessment.

Keywords:  AI; Machine Learning; Predictive models; T2D

DOI:  https://doi.org/10.1016/j.csbj.2025.06.038
Ultrasound Med Biol. 2025 Jul 12. pii: S0301-5629(25)00215-7. [Epub ahead of print]

Diabetic Tibial Neuropathy Prediction: Improving interpretability of Various Machine-Learning Models Based on Multimodal-Ultrasound Features Using SHAP Methodology.

Yanqiu Chen, Zhen Sun, Huohu Zhong, Yuwei Chen, Xiuming Wu, Liyang Su, Zhenhan Lai, Tao Zheng, Guorong Lyu, Qichen Su.

   OBJECTIVE: This study aimed to develop and evaluate eight machine learning models based on multimodal ultrasound to precisely predict of diabetic tibial neuropathy (DTN) in patients. Additionally, the SHapley Additive exPlanations(SHAP)framework was introduced to quantify the importance of each feature variable, providing a precise and noninvasive assessment tool for DTN patients, optimizing clinical management strategies, and enhancing patient prognosis.
METHODS: A prospective analysis was conducted using multimodal ultrasound and clinical data from 255 suspected DTN patients who visited the Second Affiliated Hospital of Fujian Medical University between January 2024 and November 2024. Key features were selected using Least Absolute Shrinkage and Selection Operator (LASSO) regression. Predictive models were constructed using Extreme Gradient Boosting (XGB), Logistic Regression, Support Vector Machines, k-Nearest Neighbors, Random Forest, Decision Tree, Naïve Bayes, and Neural Network. The SHAP method was employed to refine model interpretability. Furthermore, in order to verify the generalization degree of the model, this study also collected 135 patients from three other tertiary hospitals for external test.
RESULTS: LASSO regression identified Echo intensity(EI), Cross-sectional area (CSA), Mean elasticity value(Emean), Superb microvascular imaging(SMI), and History of smoking were key features for DTN prediction. The XGB model achieved an Area Under the Curve (AUC) of 0.94, 0.83 and 0.79 in the training, internal test and external test sets, respectively. SHAP analysis highlighted the ranking significance of EI, CSA, Emean, SMI, and History of smoking. Personalized prediction explanations provided by theSHAP values demonstrated the contribution of each feature to the final prediction, and enhancing model interpretability. Furthermore, decision plots depicted how different features influenced mispredictions, thereby facilitating further model optimization or feature adjustment.
CONCLUSION: This study proposed a DTN prediction model based on machine-learning algorithms applied to multimodal ultrasound data. The results indicated the superior performance of the XGB model and its interpretability was enhanced using SHAP analysis. This cost-effective and user-friendly approach provides potential support for personalized treatment and precision medicine for DTN.

Keywords:  Diabetic peripheral neuropathy; Diabetic tibial neuropathy; Machine learning; Multimodal ultrasound; Prediction model; Type 2 diabetes mellitus

DOI:  https://doi.org/10.1016/j.ultrasmedbio.2025.06.016
Diabetes Obes Metab. 2025 Jul 17.

Prediction of glycaemic control and quality of life in people with type 2 diabetes using glucose-lowering drugs with machine learning-The Maastricht study.

Nikki C C Werkman, Johannes T H Nielen, José Tapia-Galisteo, Francisco J Somolinos-Simón, Maria Elena Hernando, Junfeng Wang, Li Jiu, Wim G Goettsch, Hans Bosma, Miranda T Schram, Marleen M J van Greevenbroek, Anke Wesselius, Coen D A Stehouwer, Johanna H M Driessen, Gema Garcia-Sáez.

   BACKGROUND: Despite the heterogeneity of type 2 diabetes (T2D), all patients are treated according to the same guideline. Some people have more difficulty reaching treatment goals (adequate glycaemic control) and maintaining quality of life (QoL). Therefore, a prediction model identifying who is unlikely to reach these goals within the next year would be useful to allow specific attention to these people.
AIM: To investigate if machine learning algorithms can predict which individuals are unlikely to reach glycaemic control and likely to deteriorate in QoL in 1 year.
METHODS: We used data from The Maastricht Study, including 842 people with T2D and information on HbA1c values, and 964 people with T2D and information on QoL. We evaluated several machine learning algorithms with feature selection methods and hyperparameter tuning in fivefold cross-validation for the corresponding outcomes.
RESULTS: The prediction of inadequate glycaemic control showed good performance. The support vector machine classifier performed best in terms of accuracy (0.76 (95% CI 0.71-0.79)), precision (0.79 (95% CI 0.71-0.83)) and area under the receiver operating characteristic curve (AUC-ROC) (0.85 (95% CI 0.80-0.89)). The multi-layer perceptron classifier performed best in terms of recall (0.72 (95% CI 0.64-0.79)) and F1-score (0.73 (95% CI 0.64-0.79)). The prediction of deterioration in QoL showed inadequate performance and did not seem feasible.
CONCLUSION: Prediction of glycaemic control after 1 year in T2D is feasible with good model performance. However, the prediction of deterioration in QoL remains a challenge and needs further work.

Keywords:  antidiabetic drug; glycaemic control; observational study; type 2 diabetes

DOI:  https://doi.org/10.1111/dom.16598
J Occup Environ Med. 2025 Jul 15.

Occupational and Physical Activity Factors as Predictors of Prediabetes: A Machine Learning Study with SHAP-Based Interpretation.

Jun-Hee Kim Jh Kim.

   OBJECTIVE: A prediabetes prediction model based on nonclinical occupational and physical activity factors was developed to interpret key predictors for early intervention.
METHODS: Data from 33,265 individuals in the Korea National Health and Nutrition Examination Survey were analyzed using logistic regression, decision tree, XGBoost, and random forest algorithms. SHapley Additive exPlanations was employed for model interpretation.
RESULTS: The random forest model demonstrated the best performance (accuracy: 0.8037; AUC: 0.8597). Professional occupation, decreased average working hours, and increased walking days were associated with reduced prediabetes risk, while long working hours and high physical demands were associated with increased prediabetes risk.
CONCLUSION: Occupational and physical activity factors are important nonclinical predictors of prediabetes. These findings support targeted and accessible prevention strategies in settings with limited clinical resources, highlighting the importance of lifestyle in managing diabetes risk.

Keywords:  Machine Learning; Occupational Health; Physical Activity; Prediabetes; Prediction

DOI:  https://doi.org/10.1097/JOM.0000000000003504
PLoS One. 2025 ;20(7): e0328253

DNA sequence classification for diabetes mellitus using NuSVC and XGBoost: A comparative.

Said A Salloum, Khaled Mohammad Alomari, Ayham Salloum.

Diabetes Mellitus is a global health concern, characterized by high blood sugar levels over a prolonged period, leading to severe complications if left unmanaged. The early identification of individuals at risk is critical for effective intervention and treatment. Traditional diagnostic methods rely heavily on clinical symptoms and biochemical tests, which may not capture the underlying genetic predispositions. With the advent of genomics, DNA sequence analysis has emerged as a promising approach to uncover the genetic markers associated with Diabetes Mellitus. However, the challenge lies in accurately classifying DNA sequences to predict susceptibility to the disease, given the complex nature of genetic data. This study addresses this challenge by employing two advanced machine learning models, NuSVC (Nu-Support Vector Classification) and XGBoost (Extreme Gradient Boosting), to classify DNA sequences related to Diabetes Mellitus. The dataset, obtained from reputable sources like NCBI, was preprocessed using Natural Language Processing (NLP) techniques, where DNA sequences were treated as textual data and transformed into numerical features using TF-IDF (Term Frequency-Inverse Document Frequency). To handle the class imbalance in the dataset, SMOTE (Synthetic Minority Over-sampling Technique) was applied. The models were trained and validated using 10-fold cross-validation. XGBoost was trained with up to 300 boosting rounds, and performance was evaluated using accuracy, precision, recall, F1-score, ROC-AUC, and log loss. The results demonstrate that XGBoost outperformed NuSVC across all metrics, achieving an accuracy of 98%, a log loss of 0.0650, and an AUC of 1.00, compared to NuSVC's accuracy of 87%, log loss of 0.2649, and AUC of 0.95. The superior performance of XGBoost indicates its robustness in handling complex genetic data and its potential utility in clinical applications for early diagnosis of Diabetes Mellitus. The findings of this study underscore the importance of advanced machine learning techniques in genomics and suggest that integrating such models into healthcare systems could significantly enhance predictive diagnostics.

DOI: https://doi.org/10.1371/journal.pone.0328253
J Pharm Anal. 2025 Jun;15(6): 101305

Advancement of artificial intelligence based treatment strategy in type 2 diabetes: A critical update.

Aniruddha Sen, Palani Selvam Mohanraj, Vijaya Laxmi, Sumel Ashique, Rajalakshimi Vasudevan, Afaf Aldahish, Anupriya Velu, Arani Das, Iman Ehsan, Anas Islam, Sabina Yasmin, Mohammad Yousuf Ansari.

  In the unrelenting race to strive to dominate type 2 diabetes mellitus (T2DM) care better, this review paper sets out on a significant discovery trip across recent advancements in treatment and the blooming era of artificial intelligence (AI) utilities. Given the considerable global burden of T2DM, innovative therapeutic approaches to improve patient outcomes remain a public health priority. This review first provides an in-depth analysis of the current state of therapy, from novel pharmacotherapy to lifestyle interventions and new treatment methods. At the same time, the rapidly increasing role of AI in diabetes care is woven into the story, mainly targeting how insulin therapy can be modified and personalized through algorithms and predictive modelling. It leaves a deep review of their pre-existing synergies, which helps understand how collaborative opportunities will unlock the future of T2DM care. This critical role is shown by integrating recent therapeutic advances and AI with overall showcasing better screening, diagnosis, and therapeutics decision-making to outcome prediction in T2DM. The review emphasizes how AI applications in insulin therapy have transformative potential in diabetes care. These person-centred approaches to T2DM management, which are more effective and personalized than some traditional strategies, only work because of the often-hidden synergies between AI algorithms in areas such as diagnostic criteria, predictive methods, and familiar classification tools for subgroups with relevant aspects/predictors on prognosis or treatment responsiveness.

Keywords:  Artificial intelligence; Personalized therapy; Therapeutics; Type 2 diabetes mellitus

DOI:  https://doi.org/10.1016/j.jpha.2025.101305
Sci Rep. 2025 Jul 16. 15(1): 25706

Artificial intelligence-based diabetes risk prediction from longitudinal DXA bone measurements.

Sulaiman Khan, Zubair Shah.

Diabetes mellitus (DM) is a serious global health concern that poses a significant threat to human life. Beyond its direct impact, diabetes substantially increases the risk of developing severe complications such as hypertension, cardiovascular disease, and musculoskeletal disorders like arthritis and osteoporosis. The field of diabetes classification has advanced significantly with the use of diverse data modalities and sophisticated tools to identify individuals or groups as diabetic. But the task of predicting diabetes prior to its onset, particularly through the use of longitudinal multi-modal data, remains relatively underexplored. To better understand the risk factors associated with diabetes development among Qatari adults, this longitudinal research aims to investigate dual-energy X-ray absorptiometry (DXA)-derived whole-body and regional bone composition measures as potential predictors of diabetes onset. We proposed a case-control retrospective study, with a total of 1,382 participants contains 725 male participants (cases: 146, control: 579) and 657 female participants (case: 133, control: 524). We excluded participants with incomplete data points. To handle class imbalance, we augmented our data using Synthetic Minority Over-sampling Technique (SMOTE) and SMOTEENN (SMOTE with Edited Nearest Neighbors), and to further investigated the association between bones data features and diabetes status, we employed ANOVA analytical method. For diabetes onset prediction, we employed both conventional and deep learning (DL) models to predict risk factors associated with diabetes in Qatari adults. We used SHAP and probabilistic methods to investigate the association of identified risk factors with diabetes. During experimental analysis, we found that bone mineral density (BMD), bone mineral contents (BMC) in the hip, femoral neck, troch area, and lumbar spine showed an upward trend in diabetic patients with [Formula: see text]. Meanwhile, we found that patients with abnormal glucose metabolism had increased wards BMD and BMC with low Z-score compared to healthy participants. Consequently, it shows that the diabetic group has superior bone health than the control group in the cohort, because they exhibit higher BMD, muscle mass, and bone area across most body regions. Moreover, in the age group distribution analysis, we found that the diabetes prediction rate was higher among healthy participants in the younger age group 20-40 years. But as the age range increased, the model predictions became more accurate for diabetic participants, especially in the older age group 56-69 years. It is also observed that male participants demonstrated a higher susceptibility to diabetes onset compared to female participants. Shallow models outperformed the DL models by presenting improved accuracy (91.08%), AUROC (96%), and recall values (91%). This pivotal approach utilizing DXA scans highlights significant potential for the rapid and minimally invasive early detection of diabetes.

DOI: https://doi.org/10.1038/s41598-025-10136-5