Researchers at Rice University and the University of Texas MD Anderson Cancer Center have developed a miniature imaging device powered by artificial intelligence that could change the way clinicians detect cancer. The technology aims to bring high-resolution, real-time diagnostics directly to the point of care and was recently described in a paper published in the journal Proceedings of the National Academy of Sciences.
The device, called PrecisionView, is a handheld endoscope that overcomes the limitations of medical image processing by combining advanced optical technology and deep learning. This system allows clinicians to visualize both intracellular structures and underlying blood vessels over large tissue areas without the need for invasive biopsies.
“Early detection is one of the most important factors in improving cancer outcomes, but today’s tools often force clinicians to choose between detail and comprehensiveness,” said corresponding author Rebecca Richards Cottam, Malcolm Gillis Professor at Rice and co-director of the Rice360 Global Health Technology Institute. “With PrecisionView, we no longer have to make those tradeoffs and can see both clearly and in real time.”
Epithelial cancers, including cervical and oral cancers, account for the majority of cancer cases and are often diagnosed at late stages, in part because current diagnostic methods rely heavily on invasive and limited-scope biopsies.
Although traditional intravital microscopy provides a noninvasive alternative, it has several important limitations, including a narrow field of view, shallow depth of field, and difficulty in imaging rough tissue surfaces. These limitations can make it difficult to evaluate large or complex lesions and identify where a biopsy is truly needed. PrecisionView aims to address these challenges through a novel design that integrates a deep learning-optimized optical system and real-time image reconstruction.
The pen-sized PrecisionView uses custom-designed phase masks and AI reconstruction algorithms to dramatically expand your imaging capabilities. It provides approximately 5 times the field of view and approximately 8 times the depth of field compared to conventional systems while maintaining cellular-level resolution.
“Traditionally, machine learning and artificial intelligence tools have been used to enhance images in terms of resolution and contrast after they have been acquired with traditional imaging systems,” explained Ashok Viraraghavan, chair of Rice University’s Department of Electrical and Computer Engineering and co-author of the study. “In stark contrast, this study utilizes an AI approach to redesign the microscope optics. The AI-designed optics not only improve resolution/contrast, but more importantly, break the traditional trade-off between depth of field and resolution, creating a handheld microscope platform that achieves cellular resolution while increasing depth of field by a factor of 8. In fact, this DOF This improvement is important for the ease of use of the device in the field, as it makes it practical for clinicians and physicians.” Technicians can hold the device and obtain high-resolution images without compromising image quality due to blurring of focus. ”
This advancement allows clinicians to simultaneously visualize two important hallmarks of cancer: cellular changes in the epithelial tissue and subsurface microvascular patterns.
“Being able to capture both nuclear and vascular features in a single sequence of images is a major step forward, as these are the signals that clinicians can rely on to distinguish healthy tissue from precancerous or cancerous lesions,” said Huayu Hou, a graduate student in the Richards Courtum Optical Spectroscopic Imaging Laboratory and one of the paper’s authors.
The device produces detailed maps of tissue areas spanning several square centimeters and can display results in real time at up to 15 frames per second.
The researchers validated PrecisionView through a series of experiments, including imaging healthy volunteers and human tissue samples with precancerous lesions. In one study, the device was used to scan the mouths of volunteers, creating high-resolution maps of tissue structures and blood vessels over an area of more than 1 square centimeter. In another example, we successfully identified precancerous changes in cervical tissue, clearly distinguishing between abnormal areas and surrounding healthy tissue.
“Instead of sampling small pieces of tissue and sending them to the lab, this technology allows us to assess a much larger area instantly,” said Jimin Wu, a postdoctoral fellow in electrical and computer engineering and one of the study authors. “This could significantly reduce missed diagnoses and unnecessary procedures.”
PrecisionView is designed not only for image processing performance, but also for accessibility. The system is built using relatively simple components and costs approximately $3,000, allowing it to be deployed in clinics and low-resource settings where traditional pathology infrastructure is limited. Developing highly effective and low-cost healthcare solutions like this is one of Rice360’s signature efforts.
“PrecisionView has the potential to bring high-quality diagnostic capabilities directly to the point of care, empowering clinicians to make more timely decisions and increasing access to life-saving early detection,” said Kathleen Schmeler, one of the study’s authors and associate director of Global Oncology in the Cancer Network Division of MD Anderson Surgery. “The impact will be particularly acute in underserved areas, where access to pathology services may be limited or delayed, leading to missed or delayed diagnoses.”
Researchers say the technology could support a wide range of clinical applications, from guiding biopsy and surgical decisions to enabling early cancer detection during routine screening. However, they stress that large-scale clinical studies are still needed to fully validate the diagnostic accuracy of this device.
“PrecisionView represents the future direction of medical imaging, where artificial intelligence and optical design work together to improve outcomes,” said Richards-Kortum. “By designing hardware and algorithms together, we can unlock capabilities that were previously impossible.”
This article has been reprinted from the following material: Note: Materials may be edited for length and content. Please contact the citation source for details. You can access our press release publishing policy here.
