Biopsies remain the standard procedure for definitive diagnosis of cancer, and millions are performed annually in the U.S., but a new AI-powered handheld imaging device developed by researchers at Rice University and The University of Texas MD Anderson Cancer Center may eventually make that procedure unnecessary in many cases.
The device, called PrecisionView, is a handheld endomicroscope that enables clinicians to visualize both subcellular structures and underlying blood vessels across large tissue areas without the need for invasive biopsies. The device resembles a thick pen, but inside is a complex imaging system that combines optics with deep-learning algorithms. The technology allows doctors to examine tissue at the cellular level while also scanning a much larger area than conventional imaging systems can manage.
“Early detection is one of the most critical factors in improving cancer outcomes, but today’s tools often force clinicians to choose between detail and coverage,” said Rebecca Richards-Kortum, the Malcolm Gillis University Professor at Rice and co-director of the Rice360 Institute for Global Health Technologies. “With PrecisionView, we no longer have to make that trade-off — we can see both clearly and in real time.” Professor Richards-Kortum holds a Ph.D. in medical physics from MIT, and leads the Richards-Kortum Optical Spectroscopy and Imaging Laboratory at Rice.
Researchers from both universities collaborated to develop PrecisionView, and revealed their findings in a research paper titled “Deep-Learning endomicroscope with large field-of-view and depth-of-field for real-time in vivo imaging of epithelial cancer hallmarks.” The paper was published in the Proceedings of the National Academy of Sciences.
Epithelial cancers are the most common type of cancer, accounting for 90% of all diagnoses, developing in the thin layer of tissue that forms the lining of the skin, internal organs, and inner passageways of the body, yet many are diagnosed at late stages in part because current diagnosis relies heavily on biopsies.
“Being able to capture both nuclear and vascular features in a single, continuous image is a major step forward, because these are the signals clinicians rely on to distinguish healthy tissue from precancerous or cancerous lesions,” said Huayu Hou, one of the paper’s authors. He is a graduate student in Richards-Kortum’s laboratory, which has a mission to develop “cost effective optical imaging and spectroscopy tools to reduce the incidence and mortality of cancer and infectious disease through early detection at the point-of-care.”
The PrecisionView system achieves this using what researchers describe as a deep learning-optimized optical design.
“Traditionally, machine learning and artificial intelligence tools are used to enhance images in terms of resolution or contrast, after the images have been acquired by conventional imaging systems,” said Ashok Veeraraghavan, chair of electrical and computer engineering at Rice and a co-author of the study. “In stark contrast, this work utilizes AI approaches to redesign the optics of a microscope,” he said. “The AI-designed optics not only improves resolution/contrast but more importantly breaks the conventional trade-off between depth of field and resolution.”
Current high-resolution microscopes can easily lose focus if tissue surfaces are uneven or if the device shifts slightly in a clinician’s hand. PrecisionView’s expanded depth of field helps maintain sharp imaging even under less-than-perfect conditions.
The microscope also operates in real time, displaying reconstructed images at up to 15 frames per second. Instead of waiting days for laboratory analysis, physicians could potentially evaluate suspicious tissue during an examination.
Researchers tested the device extensively.
The system was first validated using laboratory targets, animal tissue and postmortem human breast tissue. Researchers then conducted imaging studies involving healthy volunteers and patients undergoing procedures for cervical precancer.
In one study, researchers scanned volunteers’ oral cavities, creating detailed maps of tissue structures and blood vessels across areas larger than 1 square centimeter. In another, the device successfully identified abnormal features in cervical tissue specimens containing precancerous lesions.
“Instead of sampling a small piece of tissue and sending it to a lab, this technology allows us to assess a much larger area instantly,” said Jimin Wu, one of the study’s authors. “That could significantly reduce missed diagnoses and unnecessary procedures.”
PrecisionView is built from relatively simple components and has an estimated cost of roughly $3,000, dramatically less expensive than many advanced medical imaging systems.
“PrecisionView has the potential to bring high-quality diagnostic capability directly to the point of care, helping clinicians make more timely decisions which will improve access to life-saving early detection,” said Kathleen Schmeler, associate vice president of global oncology in MD Anderson’s Division of Surgery.
“The impact will be particularly significant in medically underserved areas where access to pathology services may be limited or delayed, leading to missed or late diagnoses,” she said.
The technology is not ready to replace biopsies entirely. Researchers emphasized that larger clinical studies are still needed to confirm the device’s diagnostic accuracy across broader patient populations and multiple cancer types.
“PrecisionView represents a future direction for medical imaging, one where artificial intelligence and optical design work together to improve outcomes,” Richards-Kortum said. “By designing hardware and algorithms together, we can unlock capabilities that simply weren’t possible before.”
