Dhiya Al-Jumeily

Disease diagnosis is a critical determinant of clinical outcomes, particularly in the context of non-communicable diseases and multimorbidity, where delayed detection is associated with increased morbidity and healthcare burden. In many settings, especially under-resourced and fragmented healthcare systems, late diagnosis is often driven by limited access to affordable diagnostic tools and specialist expertise. Recent advances in artificial intelligence (AI), including deep learning and generative models, have demonstrated significant potential in extracting clinically relevant patterns from complex medical datasets, such as imaging and multimodal patient data. However, the translation of these technologies into sustainable healthcare practice remains constrained by gaps in data availability, validation, and workforce readiness. The proposed approach entails a human-in-the-loop, education-integrated framework that embeds AI-driven diagnostic research within higher education systems. The framework encompasses (i) structured real-world data collection and validation, (ii) guided development of synthetic datasets to address data scarcity, (iii) the design and evaluation of AI diagnostic models, and (iv) the integration of AI outputs into clinical decision-making processes and medical guidelines. By positioning universities as hubs for AI-enabled healthcare innovation, this approach supports the co-creation of diagnostic tools that combine algorithmic scalability with human expertise. The proposed model contributes to sustainable and cost-effective healthcare systems while fostering research capacity, workforce development, and innovation ecosystems in the Arab world. It further contributes to AI-driven economic development by enabling cost-efficient diagnostic innovation, reducing healthcare expenditure, and supporting knowledge-based economies in the region.