AI-Assisted Healthcare Kiosk: Offline Disease Prediction, Medicine Recommendation, and Telemedicine Access for Rural India

Clinical ML — AI Drug & Diet Recommendation System

The ClinicalML is an artificial intelligence-based healthcare system designed to assist in early disease prediction and personalized treatment support. It uses advanced machine learning algorithms to analyze patient data and identify health conditions accurately. The system focuses on common diseases such as diabetes, hypertension, and cardiovascular disorders using important parameters like age, BMI, blood pressure, and glucose levels. ClinicalML follows a two-stage approach in which Random Forest, Gradient Boosting, and Multi-Layer Perceptron (MLP) models are combined using a soft voting ensemble method for accurate disease classification. Based on the predicted disease, the system provides personalized diet plans and medication schedules. The complete system runs within a web browser without requiring backend servers or cloud infrastructure, ensuring faster processing, better privacy, and easy accessibility. ClinicalML also includes charts and graphs to help users and doctors understand the results and make informed healthcare decisions quickly. Keywords: AI in Healthcare, Disease Prediction, Diet & Drug Recommendation, Random Forest, Gradient Boosting, Multi-Layer Perceptron, Soft Voting Ensemble

AI Based Disease Prediction and Remedy Recommendation System

AI and ML are being utilized for enhanced treatment, diagnosis, and prediction of diseases. However, most of the applications provide generalized treatment based on the symptoms provided with lesser emphasis on patient’s health factors. In this paper, we present an AI-based system for predicting diseases along with recommending remedies. This is achieved by employing Random Forest algorithm and considering the patients’ symptoms in addition to age, gender, blood pressure, blood sugar, cholesterol, BMI, heart rate, hemoglobin, stress, and sleep duration. Further, the suggested information includes the causes, diagnostic tests, diet chart, exercise, lifestyle, home remedies, prevention measures, and specialist consultation. The proposed application has been implemented as a web application with custom API and stack, user authentication, language translation, electronic health records, admin panel, and PDF report generation. The experimental results show that the Random Forest algorithm is effective in fitting the data with accurate prediction of diseases. The proposed AI-based tool can be considered as an intelligent companion for screening diseases with improved awareness and prediction for taking necessary precautions and seeking medical treatment.

Abstract A036: Smart Health Screening in Identification of Individuals at High Cancer Risk Using Artificial Intelligence

Cancer is the second leading cause of death worldwide and causes nearly 9.6 million deaths each year according to the WHO. Although medical technology and screening programs continue to improve, delayed diagnosis remains a major challenge. Artificial intelligence (AI) offers a promising approach for cancer detection and risk prediction. This study developed an AI-based platform that integrates gene panel data, traditional risk factors, and healthcare professional assessments to identify individuals at high cancer risk. The system applied machine learning models including random forest, support vector machines, convolutional neural networks (CNN), deep learning algorithms, and Natural Language Processing to analyze complex associations between genetic variants and clinical risk factors. The platform used Python programming with Django, MySQL, ChartJS, JSCharting, HTML, CSS, and JavaScript frameworks. Developers created independent front-end and back-end modules and managed version control with Git. Researchers evaluated the platform in 514,506 cases across 39 multicenter units. The deep learning engine analyzed medical data and risk factors using CNN, support vector machines, and random forest classifiers. The platform also integrated ECG and mammography analysis with general health screening to provide comprehensive risk assessment. Two CNN models supported colon cancer analysis with different fully connected layers for classification of colon adenocarcinoma and benign colonic tissue (https://smartcancer.ir/). Random Forest-based models estimated short-term 5-year risk and long-term lifetime risk. The platform integrated Electronic Health Records (EHR) with blockchain technology to improve security, privacy, and data integrity. The system also generated family trees to support genetic counselling. The platform recommended the next clinical step based on low-risk or high-risk status and stored previous records for specialist follow-up. This study presents a novel framework that combines multiple screening modalities with AI-driven digital analysis for identification of high-risk individuals. Integration of federated learning, blockchain security, and AI-based diagnostics can improve cancer screening, support clinical decision-making, reduce cancer risk, and optimize healthcare resources. Aida Yavari Kondori, Ahmadreza Tavasouli, Mona Maftouh, Ghazaleh Pourali, Zahra Yousefli, Soodabeh Shahidsales, Ali Alamdaran, Masoud Pezeshkirad, Marjaneh Farazestanian, Elisa Giovannetti, Amir Avan, Hamid Naderi. Smart Health Screening in Identification of Individuals at High Cancer Risk Using Artificial Intelligence [abstract]. In: Proceedings of AACR Drug Discovery and Development (AACR D3) Conference; 2026 Jul 21-24; Boston, MA. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(14_Suppl):Abstract nr A036.

Harmony Search-Powered Stacked Model For Heart Disease Prediction

Adaptive Multi-Model Fusion Framework for Heart Attack and Diabetes Risk Prediction

The increase in the prevalence of diabetes and cardiovascular diseases in the world requires accurate and easy to access diagnostics. These chronic diseases are asymptomatic and need early screening in an attempt to prevent complications and reduce healthcare expenditure. The contemporary systems yield tremendous data about patients, their demographics, lifestyle, medical history, lab results, and ECGs, and the traditional tools are processing these data separately. Such a fragmented approach lacks the ability to capture complex risk interactions and predictively forecast the likelihood, slowing down intervention, and putting strain on resources in high-volume or constrained environments. Our system solves this by providing a full preprocessing pipeline: data cleaning, feature scaling, temporal slicing, and missing value imputation of consistent and reliable inputs. It uses a hybrid modelling model based on XGBoost and TabNet with a rule-based logic module with clinical heuristics. The Multimodal Transformer Fusion layer is central, as it dynamically weighted and combined predictions across all modalities so that cross-model interactions could be used to provide context-related outputs. This provides subtle patient particular predictions, enhancing accuracy of diagnosis in profiles. One of the best is the mechanism of clinician-in-the-loop feedback. It provides SAP and SHAP explainability visuals, post-prediction, which clearly explain the contribution of features. Clinicians perfect the model based on practical knowledge, which includes AI and human intelligence that increases reliability and minimizes human supervision. Focused on clinicians, diagnostic laboratories, health professionals, and health-tech innovators, this explainable and easy-to-use platform allows detecting the disease at an early stage and designing a unique treatment regimen.

AI health assistant combining transformers and XGBoost for multilingual care

CIMAS HealthMate is proposed, a hybrid multilingual VHA that integrates transformer-based natural language processing (NLP) with an explainable extreme gradient boosting (XGBoost) decision model to provide accurate and transparent symptom triage.