A Quantum-Driven Deep Reinforcement Learning Framework for Accurate Sickle Cell Disease Diagnosis and Triage

Anemia results from reduced red blood cell or hemoglobin levels, limiting oxygen transport. Sickle Cell Anemia arises from HbS polymerization under low oxygen, causing sickled cells, vaso-occlusion, hemolysis, and multi-organ complications with symptoms including pain, fatigue, and infection risk. Clinicians face challenges such as misdiagnosis, time pressure, and the need for rapid, accurate results, while conventional methods like blood smears and electrophoresis remain slow and variable. To address this, the work proposes a hybrid pipeline using DenseNet-121 for feature extraction, a Quantum Convolutional Neural Network for quantum-enhanced classification, and a Deep Q-Network for personalized clinical recommendations. The pipeline supports automated image interpretation and risk-aware decision support. Evaluated using accuracy, precision, recall, F1-score, and AUC-ROC, the model achieves 92% accuracy, demonstrating strong diagnostic capability overall. Future work includes extending to multi-class blood disorders and supporting real-time inference on edge-based quantum platforms.