Purpose Indocyanine green (ICG) fluorescence imaging is increasingly used for intraoperative bowel perfusion assessment in neonatal surgery. However, existing commercial solutions are limited to track few manually selected regions of interest (ROIs) and struggle with tissue motion and occlusion. We present a novel ICG quantification framework leveraging deep learning-based point tracking for robust perfusion analysis. Methods Our system employs CoTracker3, a state-of-the-art transformer-based point tracking model, to enable tracking of both user-specified ROIs and dense sampling across the entire visible tissue region. The framework computes comprehensive perfusion metrics including <InlineEquation ID="IEq1"> <EquationSource Format="TEX">$F_{\max }$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>F</mi> <mo movablelimits="true">max</mo> </msub> </math></EquationSource> </InlineEquation>, <InlineEquation ID="IEq2"> <EquationSource Format="TEX">$T_{\max }$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>T</mi> <mo movablelimits="true">max</mo> </msub> </math></EquationSource> </InlineEquation>, ingress gradient, <InlineEquation ID="IEq3"> <EquationSource Format="TEX">$T_{1/2\max }$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>T</mi> <mrow> <mn>1</mn> <mo stretchy="false">/</mo> <mn>2</mn> <mo movablelimits="true">max</mo> </mrow> </msub> </math></EquationSource> </InlineEquation>, and washout rate for each tracked region. For dense tissue analysis, dynamic perfusion heatmaps are generated through spatial interpolation with concave hull boundary reconstruction. Results Preliminary validation on neonatal bowel perfusion videos demonstrates robust tracking performance under tissue motion and partial occlusion. The system successfully extracts perfusion metrics comparable to the commercial PerfusionTech system while enabling dense spatial perfusion mapping not available in existing solutions. Conclusion The proposed CoTracker3-based framework provides a feasible approach for quantitative ICG fluorescence analysis with improved tracking robustness and comprehensive spatial perfusion visualization, warranting further clinical validation studies.

Robust quantification of ICG fluorescence perfusion in neonatal bowel surgery via deep point tracking

期刊论文
出版模式：

开放获取
发表日期：

2026年6月16日

查看全文

Zhehua Mao,
Antonio Composto,
Jonathan Neville,
Cecilia Cirelli,
Danail Stoyanov,
Stefano Giuliani,
Sophia Bano

摘要

Purpose

Indocyanine green (ICG) fluorescence imaging is increasingly used for intraoperative bowel perfusion assessment in neonatal surgery. However, existing commercial solutions are limited to track few manually selected regions of interest (ROIs) and struggle with tissue motion and occlusion. We present a novel ICG quantification framework leveraging deep learning-based point tracking for robust perfusion analysis.

Methods

Our system employs CoTracker3, a state-of-the-art transformer-based point tracking model, to enable tracking of both user-specified ROIs and dense sampling across the entire visible tissue region. The framework computes comprehensive perfusion metrics including $F_{\max }$ $F_{max}$ , $T_{\max }$ $T_{max}$ , ingress gradient, $T_{1/2\max }$ $T_{1 / 2 max}$ , and washout rate for each tracked region. For dense tissue analysis, dynamic perfusion heatmaps are generated through spatial interpolation with concave hull boundary reconstruction.

Results

Preliminary validation on neonatal bowel perfusion videos demonstrates robust tracking performance under tissue motion and partial occlusion. The system successfully extracts perfusion metrics comparable to the commercial PerfusionTech system while enabling dense spatial perfusion mapping not available in existing solutions.

Conclusion

The proposed CoTracker3-based framework provides a feasible approach for quantitative ICG fluorescence analysis with improved tracking robustness and comprehensive spatial perfusion visualization, warranting further clinical validation studies.