Diabetic retinopathy (DR) remains one of the leading causes of preventable blindness globally, particularly among individuals with long-standing diabetes. Early detection through regular eye examinations is essential to prevent irreversible vision loss associated with advanced stages, such as proliferative diabetic retinopathy and diabetic macular oedema. Although screening programs have been successfully deployed in various healthcare systems, rising diabetes prevalence places a growing strain on medical infrastructure. As a result, there is a critical need for scalable, automated diagnostic tools. Recent advances in artificial intelligence, particularly deep learning using convolutional neural networks (CNNs), offer promising solutions for automated analysis of retinal images. These models have demonstrated high diagnostic performance in identifying DR stages and detecting macular oedema in imaging modalities like optical coherence tomography (OCT). Several AI algorithms have now received regulatory approval and are gradually being adopted in clinical workflows. Furthermore, innovations in portable imaging devices open new avenues for patient-led monitoring and remote diagnostics. However, despite their potential, current mobile imaging systems often fall short in achieving the resolution and consistency required for reliable DR detection when compared to standard fundus photography. Integration into telemedicine platforms could bridge this gap by enabling remote screening and centralized analysis, yet real-world implementation remains limited. Challenges such as legal regulations, software interoperability, and misalignment with existing national screening protocols continue to hinder widespread adoption. This paper explores the current state of AI-assisted diabetic retinopathy screening, evaluates the readiness of emerging technologies, and discusses key barriers that must be addressed to enable global deployment and improve patient outcomes.
Background/Objectives: Colorectal cancer is a leading cause of cancer-related mortality worldwide, and early detection of gastrointestinal (GI) polyps through endoscopy is critical for improving patient outcomes. However, access to specialist gastroenterology care remains severely limited in Federal Health Professional Shortage Areas (HPSAs), particularly for high-acuity geriatric patients. This study proposes GeriAIGastroNet, a clinically oriented deep learning framework designed to support AI-assisted tele-gastroenterology workflows in resource-limited settings, with the primary objective of enabling AI-powered risk stratification and colonoscopy referral triage for elderly patients who lack on-site gastroenterology access. Methods: The framework integrates an EfficientNet-B4 backbone with multi-scale attention fusion and a geriatric severity-aware classification head to enable accurate GI polyp segmentation and automated clinical risk stratification from endoscopic images. Patients identified as high-risk are referred to colonoscopy-capable centers; such centers typically offer diagnostic colonoscopy with polypectomy capability for smaller and intermediate-complexity polyps, while patients with larger, sessile, or morphologically complex lesions requiring advanced endoscopic resection (e.g., endoscopic mucosal resection or endoscopic submucosal dissection) are further referred to tertiary endoscopy centers with specialized expertise. The model was trained and evaluated on the publicly available HyperKvasir dataset (1000 annotated polyp images). Results: GeriAIGastroNet achieved a classification accuracy of 96.77%, F1-score of 96.90%, Dice coefficient of 89.18%, and Intersection over Union (IoU) of 80.80%, outperforming established baselines, including U-Net, Attention U-Net, TransUNet, and Hybrid CNN-Transformer architectures. The integrated tele-gastroenterology decision support layer enables severity-based patient triage and automated referral triggering. Conclusions: These results demonstrate the potential of AI-powered polyp analysis to strengthen equitable access to GI care by facilitating risk stratification and specialist referral in HPSAs where direct endoscopy is unavailable, making the system deployable in telehealth infrastructures serving underserved elderly populations.