AI-powered handheld microscope aims to spot cancer earlier

Researchers from Rice University and The University of Texas MD Anderson Cancer Center have developed a handheld endomicroscope called PrecisionView, which uses advanced optics and deep learning for real-time, high-resolution diagnostics.
PrecisionView enables clinicians to visualize subcellular structures and underlying blood vessels across large tissue areas without invasive biopsies.
The device, about the size of a pen, features a custom-designed phase mask and AI reconstruction algorithm to expand its imaging capabilities significantly.
PrecisionView provides a nearly five times larger field of view and an eight times greater depth of field compared to conventional systems, all while maintaining cellular-level resolution.
The endosope is designed to visually distinguish between healthy tissue and precancerous or cancerous lesions by capturing both nuclear changes and microvascular patterns in real-time.
PrecisionView is inexpensive, costing around $3,000, and has the potential to be deployed in clinics and low-resource settings with limited pathology infrastructure.
The device is expected to significantly reduce missed diagnoses and unnecessary procedures, offering an improvement over current invasive methods.
Further clinical studies are required to fully validate the diagnostic accuracy of PrecisionView in a range of clinical applications.