Synergistic acousto-optical profiling of morphodynamic features for smart detection of acute lymphoblastic leukemia

Current cytological diagnostics for acute lymphoblastic leukemia (ALL) rely on high-end, label-based instruments that are inaccessible to decentralized clinics, leaving early-stage cases undetectable. However, early recognition is crucial for timely intervention and improved prognosis. Here, we introduce an acoustic-to-optical pathway that converts the subtle biomechanical heterogeneity of leukocytes into robust optical signatures, enabling label-free, precise screening with only a compact imaging setup. Subtle mechanical heterogeneity is amplified by controlled acoustic excitation into optically measurable spatiotemporal signals, avoiding the complex requirements for directly measuring these parameters. Acoustic waves generate shear forces that deform the cell membrane, and these distinct morphological responses are observed in malignant cells by an integrated imaging system. AI-driven tracking of the space-time evolution of cellular features enables label-free identification. Clinical validation ( n = 30) achieves 96.7% accuracy (29/30) with 96.3% sensitivity (26/27) and 100% specificity (3/3). Together, these results support scalable, image-based decentralized ALL screening and motivate further clinical translation.