Hyperspectral and machine-learning-based classification of ischemic intestinal tissue

Aug 2, 2026

Upload by Skyler

Accurate intraoperative assessment of intestinal tissue viability is essential for choosing the correct extent of resection in intestinal ischemia, yet current methods rely heavily on subjective visual and clinical signs. This research develops and validates a portable hyperspectral imaging (HSI) system combined with machine learning to objectively evaluate intestinal wall viability and distinguish reversible from irreversible ischemia. Spectral acquisition is performed on rat models at 1, 6, and 12 hours, with oxygen saturation estimated via a two-wavelength algorithm. An ML pipeline using PCA features and XGBoost is trained on histologically validated tissue classes, achieving high classification accuracy and matching saturation and histology; early human testing confirms sensitivity, with further human-data training required for deployment.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Abstract

1 Introduction

Clinical need for objective intraoperative viability assessment

Limitations and risks of current visual Kerthe method

Consequences of incorrect resection decisions

RESEARCH PAPER Hyperspectral and machine-learning-based classification of ischemic intestinal tissue Valery V. Shupletsov,a Ilya A. Goryunov,a Nikita A. Adamenkov,a,b Andrian V. Mamoshin,a,c Elena V. Potapova,a Andrey V. Dunaev,a and Viktor V. Dremina,d,* aOrel State University, Research & Development Center of Biomedical Photonics, Orel, Russia bOrel Regional Clinical Hospital, Orel, Russia cThe National Medical Research Center of Surgery Named After A. Vishnevsky, Moscow, Russia dAston University, College of Engineering and Physical Sciences, Birmingham, United Kingdom | ABSTRACT. Significance: Accurate intraoperative assessment of intestinal tissue viability is critical in determining the extent of resection in cases of intestinal ischemia. Current evaluation methods are largely subjective and lack the precision required for reliable decision-making during surgery.
Aim: We aim to develop and validate a hyperspectral imaging (HSI) system combined with machine learning (ML) to objectively assess intestinal wall viability and differentiate between reversible and irreversible ischemia.
Approach: A portable HSI system was used to acquire spectral data from rat models with induced intestinal ischemia at different time points (1, 6, and 12 h) . Tissue oxygen saturation was calculated using a two-wavelength algorithm. Spectral data were classified using an ML pipeline based on principal component analysis (PCA) and the XGBoost algorithm, trained on histologically validated tissue classes.
Results: Tissue saturation decreased with prolonged ischemia (from 66% in healthy tissue to 21% after 12 h) . Classification accuracy using PCA features reached 98% for intact tissue, 95% for possibly reversible ischemia, and 97% for irreversible ischemia. Classification maps closely matched tissue saturation distributions and histological findings. Initial clinical testing confirmed the system’s sensitivity to ischemic changes in human subjects, although further training on human data is required for ML application.
Conclusions: HSI combined with ML provides an effective, non-invasive tool for real-time intraoperative assessment of intestinal viability. This approach improves the objectivity of surgical decision-making and may reduce unnecessary resections.
© The Authors. Published by SPIE under a Creative Commons Attribution 4.0 International License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI. [DOI: 10.1117/1.JBO.30.11.116001]
Keywords: hyperspectral imaging; machine learning; XGBoost; intestinal ischemia; tissue oxygen saturation
Paper 250218GR received Jul. 15, 2025; revised Sep. 19, 2025; accepted Oct. 15, 2025; published Oct.
31, 2025. | | --- | | 1 Introduction
This paper focuses on the development of a hyperspectral imaging system to assess the degree of ischemia of intestinal wall tissue to determine its viability.
According to the clinical guidelines of the World Society of Emergency Surgery, acute mesenteric ischemia is a group of diseases characterized by impaired blood supply to various segments | *Address all correspondence to Viktor V. Dremin, [v.dremin1@aston.ac.uk](v.dremin1@aston.ac.uk) Journal of Biomedical Optics 116001-1 November 2025 • Vol. 30(11) Shupletsov et al.: Hyperspectral and machine-learning-based classification. . . of the intestine, leading to ischemia and secondary inflammatory changes.1 Ischemia results in tissue hypoxia, leading to irreversible necrotic changes in all layers of the intestinal wall. Intestinal ischemia can be non-occlusive in nature and of occlusive etiology including mesenteric artery embolism (50%), mesenteric artery thrombosis (20% to 35%), or mesenteric vein thrombosis (5% to 15%). The results of current research show that acute mesenteric ischemia accounts for 0.09% to 0.2% of hospitalizations, but mortality is as high as 80% .2 When there is doubt

FAQs

Why is objective intraoperative assessment of intestinal viability critical?

Intestinal ischemia can progress to irreversible necrosis, so surgeons must decide whether resection is necessary. Subjective assessment increases the risk of perforation and peritonitis or unnecessary resection leading to short bowel syndrome.

How does the proposed system estimate tissue oxygen saturation?

The hyperspectral imaging spectral data are processed using a two-wavelength algorithm to calculate tissue oxygen saturation.

What machine-learning approach is used for classification, and what performance is reported?

The pipeline uses principal component analysis (PCA) to derive features and XGBoost for classification, trained on histologically validated tissue classes. Reported accuracy reaches 98% for intact tissue, 95% for possibly reversible ischemia, and 97% for irreversible ischemia.

Aug 2, 2026

Upload by Skyler