Synthetic Intelligence used to boost determination making throughout colorectal most cancers surgical procedure for first time
The NIR intensities were extracted by tracking the ROIs in each video, with an emphasis on the initial ICG wash-in period of 100-300 seconds. Credit: UCD Research and Innovation
A clinical research study published in the British Journal of Surgery shows that fluorescent guidance (where a fluorescent chemical compound re-emits long wavelength light) can allow a colorectal surgeon to visually and more accurately assess cancer tissue in real time during surgery using near-infrared light ( NIR) from an administered fluorophore in conjunction with artificial intelligence (AI) methods.
This study, supported by the Disruptive Technologies and Innovation Fund 2018, examined videos of 24 patient surgeries (11 with cancer). Numerous regions of interest (ROIs) from each area of the anomaly were selected for analysis from each video. The NIR intensities were extracted by tracking the ROIs in each video, with an emphasis on the initial wash-in period. The data set used for the analysis comprised 435 ROI profiles, each with 12 perfusion-characterizing features with balanced results. At the patient level, the system correctly diagnosed 19 out of 20 cancers (95%).
Prof. Ronan Cahill, Professor of Surgery at University College Dublin (UCD) and Mater Misericordiae University Hospital (MMUH), said: “Surgery plays an essential role in the treatment of more than two-thirds of all cancers, and it becomes important surgical decisions traditionally made by human visual judgments assuming a static biological field of view (field of view) during the time frame of observation (which is moments in surgery). “
“The process of uptake and release of an external substance such as drugs and contrast agents is unique in cancer tissues. Therefore, we envisioned that an approach that combined biophysics-inspired modeling and AI could analyze intraoperative changes in NIR intensities over time in Varying Tissue , which enables clinically useful lesion classification with high specificity. To translate this knowledge into an intraoperative surgical decision support tool for the first time, a real-time computer vision AI prototype for tissue tracking and categorization was developed be used. “
Also commenting on the publication of this study in the British Journal of Surgery, Professor Donal O’Shea, Department of Chemistry at RCSI University of Medicine and Health Sciences, said: “Targeted cancer imaging agents that are currently being tested adhere strictly to traditional paradigms of the world The surgical mechanisms are mainly administered systemically prior to the operation, with the operation scheduled when there is maximally stable contrast between the tumor and other tissues. However, this point in time is often unpredictable, can take a few days and false positive results can occur. Clinical Usefulness is further limited by dosage practices, planning problems, and patient-to-patient and cancer-to-cancer differences. This work instead shows a novel way and process for instant, perfect realization of drug information during surgery, which di e would greatly improve the efficiency and effectiveness of cancer treatment. ”
A new probe could help surgeons remove tumors more accurately
Indocyanine green (ICG) perfusion with artificial intelligence for intraoperative tissue classification of colon cancer. British Journal of Surgery, znaa004, doi.org/10.1093/bjs/znaa004 Provided by University College Dublin
Quote: Artificial intelligence to improve decision-making during colon cancer surgery for the first time (2020, December 17), accessed on December 17, 2020 from https://medicalxpress.com/news/2020-12-artificial-intelligence-decision-colorectal-cancer. html
This document is subject to copyright. Except for fair trade for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.