Quantum Vision Ai For Pediatric Urological Screening Using Hybrid Quantum Machine Learning

Abstract

Pediatric urological disorders require timely recognition to reduce the risk of persistent functional, reproductive, and psychosocial consequences, yet screening still depends heavily on clinician-led visual examination. This study develops and evaluates Quantum Vision AI, a child-safe intelligent screening platform that integrates non-invasive medical imaging, Hybrid Quantum Machine Learning, explainable artificial intelligence, and clinical decision support. A Design Science Research approach combined with experimental quantitative validation was used to construct the prototype. Standardized pediatric external-genital images were preprocessed and analyzed through a hybrid quantum-classical architecture in which classical vision models extracted anatomical features, quantum circuits enhanced feature representation, and explainability methods visualized diagnostically relevant regions. The proposed platform consistently outperformed the conventional convolutional, EfficientNetV2, and Vision Transformer baselines used in the study, while its attention maps concentrated on clinically meaningful anatomical structures and expert reviewers reported favorable usability and diagnostic relevance. These findings demonstrate the feasibility of quantum-enhanced computer vision for pediatric urological screening and support further multicenter validation, dataset expansion, and clinical translation. The platform is intended to augment, rather than replace, physician judgment in early screening and referral decisions.