Long-term, flexible therapy strategies are needed for chronic diseases like cardiovascular ailments, diabetes, and chronic kidney disease and they have to deal with patient heterogeneity, changing physiological states and privacy constraints. This research suggests a framework for privacy-preserving and personalized artificial intelligence that combines Federated Learning (FL), a Res-HyperTransformerNet deep prediction model, Personalized Multi-Agent Reinforcement Learning (PMARL) and Explainable Artificial Intelligence (XAI) for adaptive chronic disease management. The framework is structured with two main aims: (i) to create an accurate and privacy-aware predictive model for chronic disease risk using diverse data sources, and (ii) to permit adaptive, individual therapy optimization through multi-agent reinforcement learning. Federated learning is utilized to conduct the training of Res-HyperTransformerNet over distributed Internet of Medical Things (IoMT) nodes without the requirement of transferring raw patient data. The embeddings that are created from patient data are then passed on to a PMARL module, where several agents optimize therapy dimensions like medications, diet, physical activity, and mental health interventions independently. To make the clinical support more understandable, the SHAP-based explainability method is used for both predictive and decision-making parts. The framework is tested using two public datasets—the CDC Chronic Disease dataset and the UCI Chronic Kidney Disease Risk Factor dataset. The performance is measured by employing classification metrics (accuracy, precision, recall, F1-score, MCC) as well as reinforcement learning metrics (reward score, convergence steps, episode return) and federated system metrics (communication overhead, convergence rounds, and training time). The experiments indicate that the proposed framework demonstrated improved performance in terms of predictive accuracy and policy convergence speed compared to the baseline deep learning and reinforcement learning models, and at the same time it is more cost-effective when it comes to communication in a federated setting. This means that the proposed method suggests the potential of being a reproducible and open AI framework for adaptive chronic disease therapy management. The combination of ResNet and Transformer blocks, i.e., Federated Res-HyperTransformerNet, achieved strong performance in both datasets, namely CDC Chronic Disease (Dataset 1) and UCI CKD Risk Factor (Dataset 2). The obtained accuracy for Dataset 1 was 98.61% and for Dataset 2, it was 97.75%, under the conducted experimental settings.
Diabetes, a chronic condition affecting millions worldwide, necessitates early intervention to prevent severe complications. While accurately predicting diabetes onset or progression remains challenging due to complex and imbalanced datasets, recent advancements in machine learning offer potential solutions. Traditional prediction models, often limited by default parameters, have been superseded by more sophisticated approaches. Leveraging Bayesian optimization to fine-tune XGBoost, researchers can harness the power of complex data analysis to improve predictive accuracy. By identifying key factors influencing diabetes risk, personalized prevention strategies can be developed, ultimately enhancing patient outcomes. Successful implementation requires meticulous data management, stringent ethical considerations, and seamless integration into healthcare systems. This study focused on optimizing the hyperparameters of an XGBoost ensemble machine learning model using Bayesian optimization. Compared to grid search XGBoost (accuracy: 97.24%, F1-score: 95.72%, MCC: 81.02%), the XGBoost with Bayesian optimization achieved slightly improved performance (accuracy: 97.26%, F1-score: 95.72%, MCC:81.18%). Although the improvements observed in this study are modest, the optimized XGBoost model with Bayesian optimization represents a promising step towards revolutionizing diabetes prevention and treatment. This approach holds significant potential to improve outcomes for individuals at risk of developing diabetes.
The incidence of suicide, self-harm, and mental crises among teenagers is rising, presenting significant global public health issues. Conventional clinical risk evaluations have inadequate predictive accuracy, often overlooking high-risk adolescents. Machine learning models employing electronic health records provide an innovative method for predicting mental health crises through the integration of intricate clinical data patterns. Hence, this review aims to comprehensively examine and synthesize information about machine learning models created using electronic health record data for predicting suicide attempts, self-harm, or mental hospitalization in teenagers. Adhering to Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) principles, we executed an exhaustive search across six databases (2000-2025) for peer-reviewed research utilizing machine learning algorithms on electronic health record data to predict adolescent mental health crises. The inclusion criteria emphasized structured or unstructured electronic health record inputs, teenage cohorts (ages 10-20), and performance indicators like area under the curve (AUC), sensitivity, or specificity. The Prediction Model Risk of Bias Assessment Tool (PROBAST) was utilized to evaluate the risk of bias. Our search yielded five studies (2019-2024) that satisfied the inclusion criteria. All studies were retrospective cohorts conducted in high-income nations. Structured electronic health record data (e.g., diagnoses, prescriptions) were frequently utilized; two studies included natural language processing. Machine learning models demonstrated moderate to high discrimination (AUC 0.68-0.88), exhibiting optimal performance in short-term suicide prediction with hybrid data inputs. All investigations, however, exhibited a significant risk of bias in the analysis domain owing to insufficient external validation and absent calibration data. The positive predictive values consistently remained modest across all models. Overall, machine learning models demonstrate potential for enhancing adolescent suicide risk classification utilizing electronic health record data, surpassing numerous traditional instruments. Nonetheless, issues of generalizability, ethical constraints, and implementation obstacles remain. Thorough validation, calibration, and equity assessments are necessary prior to incorporation into clinical practice.
adolescent mental healthsuicide risk predictionself-harmelectronic health records
Plastic and reconstructive surgery increasingly depends on data-driven instruments to improve clinical decision-making. Machine learning (ML), through its capacity to analyze intricate, high-dimensional data, presents novel prospects for forecasting surgical outcomes with enhanced accuracy compared to traditional statistical models. This systematic study evaluates the extent, efficacy, and methodological rigor of ML applications in plastic and reconstructive surgery, encompassing burn treatment, microsurgical reconstruction, and breast surgery. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 standards, we searched four main databases for research published between January 2015 and March 2025. Eligible papers used ML models to predict clinical outcomes in plastic or reconstructive surgery and provided quantifiable performance indicators. Data were retrieved using the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) and Prediction Model Risk of Bias Assessment Tool (PROBAST) frameworks, then synthesized narratively. We included eleven studies that had more than 34,000 patients. Random forests, neural networks, and gradient boosting have emerged as the most prevalent and highest-performing models, with several attaining AUCs exceeding 0.90. Notwithstanding encouraging outcomes, the majority of research depended on internal validation, with only one performing external validation. Calibration reporting and data transparency were constrained. The majority of studies exhibited a high or moderate risk of bias. In summary, ML models provide significant prediction accuracy in plastic surgery; yet, they are limited by methodological deficiencies that impede clinical implementation. Future initiatives must prioritize external validation, repeatability, and ethical execution to facilitate the safe and successful incorporation into surgical practice.
Estimating antemortem blood alcohol concentration (BAC) after death is complicated by postmortem artifacts (endogenous ethanol formation, contamination, redistribution). Vitreous humor is anatomically protected and may resist these artifacts, offering a more stable matrix. The detection of ethyl glucuronide (EtG) and ethyl sulfate (EtS) can further distinguish true antemortem intake from postmortem production. Our objective is to evaluate the diagnostic reliability of vitreous ethanol concentration (VAC) as a surrogate for antemortem BAC, and to define contexts where vitreous testing adds forensic value. This review of postmortem human studies compared VAC with BAC, while adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines for systematic reviews. We searched PubMed, Scopus, Web of Science, and Google Scholar from July 2010 to July 2025, screened in duplicate, extracted data using a standardized form, and appraised study quality with a revised Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool for forensic accuracy. Given methodological and clinical heterogeneity (analytical techniques, postmortem interval (PMI), decomposition), we performed a structured narrative synthesis. Nine studies (Europe and Asia) met our inclusion criteria. Most used headspace gas chromatography with flame ionization detection (HS‑GC/FID) for ethanol; several incorporated liquid chromatography-tandem mass spectrometry (LC-MS/MS) for EtG/EtS. VAC and BAC were strongly correlated (r ≈ 0.89-0.97) under standardized sampling and shorter PMIs. Vitreous matrices showed greater resilience in cases with decomposition, long PMI, or suspected microbial ethanol formation. Studies measuring EtG/EtS indicated that combined ethanol-metabolite interpretation improves specificity for antemortem drinking and reduces false positives. Despite high correlation, individual BAC predictions from VAC had wide confidence/prediction intervals, underscoring the need for contextual, multi‑parameter interpretation. QUADAS‑2 suggested a low risk of bias in most domains across included studies. Meta‑analysis was not feasible due to heterogeneity. In conclusion, vitreous ethanol is a forensically robust adjunct or alternative to peripheral blood when BAC is unavailable or compromised, particularly with decomposition or suspected postmortem ethanol formation. Interpretation should pair VAC with EtG/EtS, femoral BAC (when available), PMI, and scene/autopsy data. Standardizing analytical protocols and validating calibrated prediction models across PMI and decomposition strata are priorities to reduce case‑level uncertainty.
BMC Medical Informatics and Decision Making · مجلد 24 (1) · 92-92
Emerging from the convergence of digital twin technology and the metaverse, consumer health (MCH) is witnessing a transformative shift. The amalgamation of bioinformatics with healthcare Big Data has ushered in a new era of disease prediction models that harness comprehensive medical data, enabling the anticipation of illnesses even before the onset of symptoms. In this model, deep neural networks stand out because they improve accuracy remarkably by increasing network depth and making weight changes using gradient descent. Nonetheless, traditional methods face their own set of challenges, including the issues of gradient instability and slow training. In this case, the Broad Learning System (BLS) stands out as a good alternative. It gets around the problems with gradient descent and lets you quickly rebuild a model through incremental learning. One problem with BLS is that it has trouble extracting complex features from complex medical data. This makes it less useful in a wide range of healthcare situations. In response to these challenges, we introduce DAE-BLS, a novel hybrid model that marries Denoising AutoEncoder (DAE) noise reduction with the efficiency of BLS. This hybrid approach excels in robust feature extraction, particularly within the intricate and multifaceted world of medical data. Validation using diverse datasets yields impressive results, with accuracies reaching as high as 98.50%. DAE-BLS's ability to rapidly adapt through incremental learning holds great promise for accurate and agile disease prediction, especially within the complex and dynamic healthcare scenarios of today.
disease predictiondigital twinmetaverseconsumer health
Monoamine oxidases (MAOs), specifically MAO-A and MAO-B, play important roles in the breakdown of monoamine neurotransmitters. Therefore, MAO inhibitors are crucial for treating various neurodegenerative disorders, including Parkinson's disease (PD), Alzheimer's disease (AD), and amyotrophic lateral sclerosis (ALS). In this study, we developed a novel cheminformatics pipeline by generating three diverse molecular feature-based machine learning-assisted quantitative structural activity relationship (ML-QSAR) models concerning MAO-B inhibition. PubChem fingerprints, substructure fingerprints, and one-dimensional (1D) and two-dimensional (2D) molecular descriptors were implemented to unravel the structural insights responsible for decoding the origin of MAO-B inhibition in 249 non-reductant molecules. Based on a random forest ML algorithm, the final PubChem fingerprint, substructure fingerprint, and 1D and 2D molecular descriptor prediction models demonstrated significant robustness, with correlation coefficients of 0.9863, 0.9796, and 0.9852, respectively. The significant features of each predictive model responsible for MAO-B inhibition were extracted using a comprehensive variance importance plot (VIP) and correlation matrix analysis. The final predictive models were further developed as a web application, MAO-B-pred ( https://mao-b-pred.streamlit.app/ ), to allow users to predict the bioactivity of molecules against MAO-B. Molecular docking and dynamics studies were conducted to gain insight into the atomic-level molecular interactions between the ligand-receptor complexes. These findings were compared with the structural features obtained from the ML-QSAR models, which supported the mechanistic understanding of the binding phenomena. The presented models have the potential to serve as tools for identifying crucial molecular characteristics for the rational design of MAO-B target inhibitors, which may be used to develop effective drugs for neurodegenerative disorders.
monoamine oxidase B inhibitorsMAO-B inhibitionPubChem fingerprintsmolecular descriptors
Bioinformatics and Healthcare Integration Disease prediction models have been revolutionized by Big Data. These models, which make use of extensive medical data, predict illnesses before symptoms appear. Deep neural networks are well-known for their ability to increase accuracy by extending the network's depth and modifying weights through gradient descent. Traditional approaches, however, are hindered by issues such as gradient instability and delayed training. As a substitute, the Broad Learning (BL) system is introduced, which avoids gradient descent in favor of quick reconstruction by incremental learning. However, BL has trouble extracting complicated features from medical data, which makes it perform poorly in cases involving complex healthcare. We suggest ABL, which combines the effectiveness of BL with the noise reduction of Denoising Auto Encoder (AE), to address this. Robust feature extraction is an area in which the hybrid model shines, especially in intricate medical environments. Accuracy of up to 98.50 % is achieved by remarkable results from validation using a variety of datasets. The ability of ABL to quickly adapt through incremental learning suggests that it may be used to forecast diseases in complicated healthcare contexts with agility and accuracy.
disease predictionbroad learningdenoising autoencoderhybrid deep neural network
Brain stroke (BS) imposes a substantial burden on healthcare systems due to the long-term care and high expenditure. Earlier detection and intervention can reduce the impact of BS. Magnetic resonance imaging (MRI) is commonly applied for BS detection. Deep learning techniques can employ MRI images to identify the BS risks in the initial stages. This study developed a BS detection model using an ensemble learning approach that combines the predictions of the base models. A MobileNet V3 model backbone was used to extract the intricate patterns of BS from MRI images. LightGBM and CatBoost models were used as base models to predict BS using the extracted features. In addition, the random forest model was used to integrate the predictions of base models to identify BS. The proposed model was generalized on a public MRI dataset that covers 2888 clinical MRI images. The experimental outcomes showed the effectiveness of the suggested BS detection model. The proposed model has obtained an accuracy of 98.7%, an area under the receiver operating characteristic score of 0.95, and an area under the precision–recall curve of 0.92. The recommended model is believed to be deployed in real-time healthcare settings to assist radiologists and clinicians in making effective decisions.
Frontiers in Molecular Biosciences · مجلد 11 · 1426964-1426964
Background: Breast cancer (BC) is a significant cause of morbidity and mortality in women. Although the important role of metabolism in the molecular pathogenesis of BC is known, there is still a need for robust metabolomic biomarkers and predictive models that will enable the detection and prognosis of BC. This study aims to identify targeted metabolomic biomarker candidates based on explainable artificial intelligence (XAI) for the specific detection of BC. Methods: Data obtained after targeted metabolomics analyses using plasma samples from BC patients (n = 102) and healthy controls (n = 99) were used. Machine learning (ML) models based on raw data were developed, then feature selection methods were applied, and the results were compared. SHapley Additive exPlanations (SHAP), an XAI method, was used to clinically explain the decisions of the optimal model in BC prediction. Results: The results revealed that variable selection increased the performance of ML models in BC classification, and the optimal model was obtained with the logistic regression (LR) classifier after support vector machine (SVM)-SHAP-based feature selection. SHAP annotations of the LR model revealed that Leucine, isoleucine, L-alloisoleucine, norleucine, and homoserine acids were the most important potential BC diagnostic biomarkers. Combining the identified metabolite markers provided robust BC classification measures with precision, recall, and specificity of 89.50%, 88.38%, and 83.67%, respectively. Conclusion: In conclusion, this study adds valuable information to the discovery of BC biomarkers and underscores the potential of targeted metabolomics-based diagnostic advances in the management of BC.
breast cancermetabolomic biomarkersplasma metabolomicsbreast cancer diagnosis
Process mining, a burgeoning discipline within data science, demonstrates a significant contribution to the software development lifecycle of diverse real-time consumer-centric projects. This paper underscores the prominence of integrating predictive business process monitoring into organizational process models, as it can substantially impact profits and efficiency in any possible business domain along with improving services to consumers. The paper proposes a novel deep learning-based business process prediction model consisting of multiple layers with fine-tuning hyperparameters. The proposed model leverages input embeddings to represent each of the activities, and based on the training of the proposed model, the accuracy of the next activity is calculated. To assess the efficacy of the proposed model, it has been compared with the existing benchmark models. Our proposed model has shown a significant gain over the existing approaches. The results show that the proposed model outperforms these approaches by achieving an accuracy of 76% on the consumer helpdesk dataset along with an accuracy of 78% on the benchmark BPI dataset.
process miningpredictive monitoringnext activity predictiondeep learning
International Conference on Contemporary Computing and Informatics (IC3I) · 1556-1562
Diagnostic prediction experts use a patient’s medical history to anticipate their future diagnosis. Attention mechanisms and recurrent neural networks are typically utilized for this purpose; however, data shortages often cause complications. Additionally, healthcare information can help diagnostic prediction algorithms. This study links healthcare ideas using AI-aware, intelligent healthcare ontologies and patient visit records as an intelligent graph. Next, utilize a graph convolution module to collect geographical attributes that show patient visit healthcare code linkages. Using a multihead attention technique, visit information and healthcare knowledge are compared to predict the patient’s future health. On two open-source healthcare datasets, the suggested solution outperforms state-of-the-art diagnostic prediction algorithms and better uses IMG data.
Hepatitis C Virus (HCV) is a viral infection that causes liver inflammation. Annually, approximately 3.4 million cases of HCV are reported worldwide. A diagnosis of HCV in earlier stages helps to save lives. In the HCV review, the authors used a single ML-based prediction model in the current research, which encounters several issues, i.e., poor accuracy, data imbalance, and overfitting. This research proposed a Hybrid Predictive Model (HPM) based on an improved random forest and support vector machine to overcome existing research limitations. The proposed model improves a random forest method by adding a bootstrapping approach. The existing RF method is enhanced by adding a bootstrapping process, which helps eliminate the tree's minor features iteratively to build a strong forest. It improves the performance of the HPM model. The proposed HPM model utilizes a 'Ranker method' to rank the dataset features and applies an IRF with SVM, selecting higher-ranked feature elements to build the prediction model. This research uses the online HCV dataset from UCI to measure the proposed model's performance. The dataset is highly imbalanced; to deal with this issue, we utilized the synthetic minority over-sampling technique (SMOTE). This research performs two experiments. The first experiment is based on data splitting methods, K-fold cross-validation, and training: testing-based splitting. The proposed method achieved an accuracy of 95.89% for k = 5 and 96.29% for k = 10; for the training and testing-based split, the proposed method achieved 91.24% for 80:20 and 92.39% for 70:30, which is the best compared to the existing SVM, MARS, RF, DT, and BGLM methods. In experiment 2, the analysis is performed using feature selection (with SMOTE and without SMOTE). The proposed method achieves an accuracy of 41.541% without SMOTE and 96.82% with SMOTE-based feature selection, which is better than existing ML methods. The experimental results prove the importance of feature selection to achieve higher accuracy in HCV research.
hepatitis C virushybrid predictive modelimproved random forestbootstrapping
A pathological complete response to neoadjuvant chemoradiotherapy offers patients with rectal cancer that has advanced locally the highest chance of survival. However, there isn't yet a valid prediction model available. An efficient feature extraction technique is also required to increase a prediction model's precision. CDAS (cancer data access system) program is a great place to look for cancer along with images or biospecimens. In this study, we look at data from the CDAS system, specifically Bowel cancer (colorectal cancer) datasets. This study suggested a survival prediction method for rectal cancer. In addition, determines which deep learning algorithm works best by comparing their performance in terms of prediction accuracy. The initial job that leads to correct findings is corpus cleansing. Moving forward, the data pre-processing activity will be performed, which will comprise "exploratory data analysis and pruning and normalization or experimental study of data, which is required to obtain data features to design the model for cancer detection at an early stage." Aside from that, the data corpus is separated into two sub-corpora: training data and test data, which will be utilized to assess the correctness of the constructed model. This study will compare our auto-encoder accuracy to that of other deep learning algorithms, such as ANN, CNN, and RBM, before implementing the suggested methodology and displaying the model's accuracy graphically after the suggested new methodology or algorithm for patients with rectal cancer. Various criteria, including true positive rate, ROC curve, and accuracy scores, are used in the experiments to determine the model's high accuracy. In the end, we determine the accuracy score for each model. The outcomes of the simulation demonstrated that rectal cancer patients may be estimated using prediction models. It is shown that variational deep encoders have excellent accuracy of 94% in this cancer prediction and 95% for ROC curve regions. The findings demonstrate that automated prediction algorithms are capable of properly estimating rectal cancer patients’ chances of survival. The best results, with 95% accuracy, were generated by deep autoencoders.