Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206587,
        author = {Hussain Hasim and Aryan Subbaiah MS and Mohammed Zimad and Deepak KM and Kavitha},
        title = {KindredHealth: An AI-Powered Multimodal Health Triage and Early Screening System},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {42-48},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206587},
        abstract = {The healthcare industry around the world faces numerous challenges in providing timely, precise, and efficient initial medical diagnosis because of a lack of specialized physicians, increasing patient numbers, and the distributed nature of clinical data. In this paper, KindredHealth is introduced, an AI-based healthcare triage and early diagnosis system.33333333 The system integrates user-reported symptom data, medical images such as skin condition photographs, and basic physiological parameters to generate preliminary diagnostic insights. A hybrid framework combining convolutional neural networks (CNNs) for image-based analysis with rule-based and ensemble machine learning models for symptom evaluation is proposed. The system predicts probable medical conditions, estimates severity levels, and recommends appropriate medical consultation pathways. Explainable AI (XAI) methods such as Grad-CAM and SHAP values are used to provide clear and understandable explanations for the model's predictions. The proposed algorithm is experimentally tested using the ISIC 2019 dermoscopic benchmark dataset as well as custom created clinical vignettes, with the results showing that KindredHealth has obtained 94.8% top-1 diagnostic accuracy when operating in multimodal fusion mode and 0.88 Cohen's kappa for urgency-classification. KindredHealth is not meant to replace doctors in any way and operates only as a decision support system which enables quicker analysis, reduces delays in diagnosis, and increases access to health care for marginalized groups.},
        keywords = {Artificial Intelligence, Clinical Decision Support, Convolutional Neural Network, Explainable AI, Health Triage, Machine Learning, Medical Image Analysis, Multimodal Diagnosis, Symptom Checker, Telemedicine.},
        month = {July},
        }