AB096. SOH25_AB_357. Iterative evolution and deployment of AUGUR-AI: a real-time AI interpreter for indocyanine green fluorescence angiography in colorectal surgery
Plenary Session

AB096. SOH25_AB_357. Iterative evolution and deployment of AUGUR-AI: a real-time AI interpreter for indocyanine green fluorescence angiography in colorectal surgery

Philip Mc Entee1, Conor Delaney2, Ashokkumar Singaravelu2, Pol Mac Aonghusa2, Ronan Cahill1

1Department of Surgery, Mater Misericordiae University Hospital, Dublin, Ireland; 2Centre for Precision Surgery, University of Dublin, Dublin, Ireland


Background: Indocyanine green fluorescence angiography (ICGFA) use intraoperatively is proven to reduce colorectal anastomotic leak rates following left-sided and rectal resections. However, its interpretation is subjective and requires experience for benefit. Building on proof-of-concept work, we have evolved and deployed a real-time artificial intelligence (AI) interpreter to provide ICGFA-decision assistance in surgery, including zero-touch user interface.

Methods: AUGUR-AI effectively represents expert interpreted ICGFA signals on recorded videos from a single ICGFA imager (PINPOINT, Stryker, USA) via breakthrough AI time-series analysis (including long short-term memory modelling). This is grounded in expert surgeon interpretation regarding stapler placement to characterise perfusion, in on-screen intestinal segments. For clinical use, seamless in-surgery deployment with other imagers is needed.

Results: The python-coded AUGUR-AI prototype was migrated to self-contained GPU (AGX Development kit, Jetson, Nvidia), and thereafter successfully deployed in-theatre with direct video feed in a series of 13 consenting patients (18 intestinal segments, including distal resection margins) undergoing colorectal resection (5 rectal resections), including alternative commercial surgical imager (1788 Stryker) (Institutional approval: 1/378/2092). In a subset of 10 patients, zero-touch intraoperative on-screen intestinal segmentation and selection was utilised, including in 3 patients by the table-side scrubbed surgeon (via boom-based webcam and hand-gesture computer vision). The system’s performance, including accuracy, was maintained >95%, enabling a usable, first-in-class deployment with meaningful and actionable output.

Conclusions: AUGUR-AI is now usable for colorectal surgeons, enabling both advanced automatic computer vision segmentation (prototype now established) and pre-CE prospective, international, multicentre validation (approved, registered (Institutional approval: 1/378/2450), and in site setup).

Keywords: Artificial intelligence(AI); indocyanine green fluorescence angiography (ICGFA); colorectal surgery; quantification; deployment


Acknowledgments

None.


Footnote

Funding: None.

Conflicts of Interest: The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.


doi: 10.21037/map-25-ab096
Cite this abstract as: Mc Entee P, Delaney C, Singaravelu A, Mac Aonghusa P, Cahill R. AB096. SOH25_AB_357. Iterative evolution and deployment of AUGUR-AI: a real-time AI interpreter for indocyanine green fluorescence angiography in colorectal surgery. Mesentery Peritoneum 2025;9:AB096.

Download Citation