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{200116,
        author = {ASHVANT NARAYAN Y and DHIVESH Y S and Mrs. R. PUSHPALATHA},
        title = {Livarix AI: A Dual-Model Deep Learning Framework for Simultaneous Liver and Lung Cancer Detection with Automated Clinical Report Generation},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {1197-1203},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=200116},
        abstract = {Cancer diagnosis continues to place significant demands on radiological infrastructure, particularly in regions where trained specialists are scarce. Manual interpretation of CT scans is time-consuming, prone to variability between observers, and difficult to scale across large patient populations. To address these challenges, this work proposes Livarix AI — a unified, end-to-end deep learning platform capable of simultaneously screening for lung and liver malignancies from CT imaging data. The lung detection component is built on an EfficientNet-B0 encoder-head architecture, which achieves a validation accuracy of 94.7% and a macro-AUC of 0.971. The liver detection component employs a Vision Transformer-Hybrid (ViT-Hybrid) model with hierarchical self-attention, attaining 91.3% validation accuracy and a macro-AUC of 0.943. Both components incorporate Gradient-weighted Class Activation Mapping (Grad-CAM) to produce spatially interpretable heatmaps for clinical transparency. The complete system is deployed as a React-based web application hosted on Vercel, with inference served via a FastAPI backend on HuggingFace Spaces. Automated clinical report generation — including TNM staging, dietary recommendations, and specialist referral pathways — is supported in both free HTML and premium PDF formats. The platform is aligned with UN Sustainable Development Goal 3 (Good Health and Well-Being) and represents a functional, deployment-ready clinical decision support system.},
        keywords = {Lung Cancer Detection, Liver Cancer Detection, EfficientNet, Vision Transformer, Grad-CAM, Deep Learning, Medical Imaging, Clinical Report Generation, Transfer Learning, Explainable AI, Computer-Aided Diagnosis, UN SDG 3},
        month = {May},
        }