D24AC00358

Surgical removal of a tumor is a mainstay treatment for many cancer types.

However, because the surgeon does not have a realtime microscopic view of the cancerous tissue during the surgical procedure, resected tissue specimens are sent to pathology for processing and visualization of the tissue and cells under a microscope either while the surgeon team and patient wait in the operating room or after the surgical procedure has ended.

Unfortunately, for many surgical oncology procedures, particularly those where the surgeon attempts to preserve as much normal tissue as possible, there are exceedingly high positive margin rates.

Finding the presence of tumor cells at or near the cut surgical margin on the resected tissue means that tumor cells have been left behind in the patient.

This leads to reoperations, added costs, time and anxiety, and an increased likelihood for recurrence of the cancer.

This project proposes an end-to-end solution to meet the needs for reducing positive margins, reoperation rates, and local recurrence rates in surgical oncology procedures.

A comprehensive team of leading experts has been assembled to develop, demonstrate, and deliver a platform solution to meet this critical need.

The team includes world-leading experts in optical imaging, robotic surgery, and AI from the University of Illinois Urbana Champaign, key knowledge and opinion leaders, and renowned physicians and experts in human factors and health economics.

Major medical technology industry partners that have proven experience at translating new surgical technologies into viable commercial products are interested in this platform technology.

The proposed technology addresses a major challenge: how can one scan and image large areas of tissue during surgery at the microscopic subcellular scale needed to make a histopathological diagnosis, all in realtime to ensure that all tumor tissue and cells have been removed?

We propose a multimodal, multiscale approach that uses optical coherence tomography (OCT) imaging of the large tumor cavity area with switchable high-resolution subcellular nonlinear optical (NLO) imaging for generating realtime digital histopathology images.

The technology captures optical biopsies entirely label-free, thereby obviating the need for administering contrast agents, questioning whether the agents are on or off their intended targets, or incurring time delays with preclinical safety evaluation and FDA approval.

Previously published studies by our team have already demonstrated intraoperative OCT of the tumor resection bed in humans and intraoperative NLO of resected tumor tissues.

AI algorithms are integrated to both rapidly screen incoming OCT images to alert the surgeon to suspicious areas, as well as to analyze NLO images to detect and diagnose for realtime classification and intraoperative decision making without having to send tissues to a pathologist.

The platform solution will utilize a handheld probe for wide position sensed manual scanning that can then be attached to a robotic arm for precision scanning in open surgical procedures for breast, lung, hepatobiliary, and pancreatic cancer.

Fiber-based beam delivery will enable flexible endoscope and rigid laparoscope options for robotic surgical implementation in the future.

By design, the platform will function for realtime in vivo intraoperative use but can also be used for assessment of ex vivo tissues as well.

We believe it is more important to determine what may be left behind in the patient rather than image what was removed.

At the end of this 5-year project period, this MARGINDX platform solution will be proven and commercially viable for rapid dissemination and use across the nation and globe, and bring a newfound standard of care for precision surgical interventions.

University Of Illinois

THE FOUR INITIAL FOCUS AREAS ARE(1) HEALTH SCIENCE FUTURES(2) SCALABLE SOLUTIONS(3) PROACTIVE HEALTH(4) RESILIENT SYSTEMS

93.384 - ADVANCED RESEARCH PROJECTS AGENCY for HEALTH (ARPA-H)

Interior Business Center (IBC) [DOI]

Illinois United States

State-Wide

D-AQD-FA-24-023