AI-Enabled Vessels Segmentation Model for Real-Time Laparoscopic Ultrasound Imaging

Ignas Kupcikevicius, Luca Boretto, Inger A. Grunbeck, Rahul Prasanna Kumar, Varatharajan Nainamalai, Mehdi Sadat Akhavi, Bjørn Edwin, Ole Jakob Elle
Proceedings of the 7th Northern Lights Deep Learning Conference (NLDL), PMLR 307:263-273, 2026.

Abstract

Laparoscopic ultrasound (LUS) is essential for assessing the liver during laparoscopic liver resections. However, the interpretation of LUS images presents significant challenges due to the steep learning curve and image noise. In this study, we propose an enhanced U-Net-based neural network with a ResNet18 backbone specifically designed for real-time liver vessel segmentation of 2D LUS images. Our approach incorporates five preprocessing steps aimed at maximizing the training information extracted from the ultrasound sonogram region. The modified U-Net model achieved a Dice coefficient of 0.879, demonstrating real-time performance at 40 frames per second and enabling the development of advanced ultrasound-based surgical navigation solutions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v307-kupcikevicius26a, title = {{AI}-Enabled Vessels Segmentation Model for Real-Time Laparoscopic Ultrasound Imaging}, author = {Kupcikevicius, Ignas and Boretto, Luca and Grunbeck, Inger A. and Kumar, Rahul Prasanna and Nainamalai, Varatharajan and Akhavi, Mehdi Sadat and Edwin, Bj{\o}rn and Elle, Ole Jakob}, booktitle = {Proceedings of the 7th Northern Lights Deep Learning Conference (NLDL)}, pages = {263--273}, year = {2026}, editor = {Kim, Hyeongji and Ramírez Rivera, Adín and Ricaud, Benjamin}, volume = {307}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jan}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v307/main/assets/kupcikevicius26a/kupcikevicius26a.pdf}, url = {https://proceedings.mlr.press/v307/kupcikevicius26a.html}, abstract = {Laparoscopic ultrasound (LUS) is essential for assessing the liver during laparoscopic liver resections. However, the interpretation of LUS images presents significant challenges due to the steep learning curve and image noise. In this study, we propose an enhanced U-Net-based neural network with a ResNet18 backbone specifically designed for real-time liver vessel segmentation of 2D LUS images. Our approach incorporates five preprocessing steps aimed at maximizing the training information extracted from the ultrasound sonogram region. The modified U-Net model achieved a Dice coefficient of 0.879, demonstrating real-time performance at 40 frames per second and enabling the development of advanced ultrasound-based surgical navigation solutions.} }
Endnote
%0 Conference Paper %T AI-Enabled Vessels Segmentation Model for Real-Time Laparoscopic Ultrasound Imaging %A Ignas Kupcikevicius %A Luca Boretto %A Inger A. Grunbeck %A Rahul Prasanna Kumar %A Varatharajan Nainamalai %A Mehdi Sadat Akhavi %A Bjørn Edwin %A Ole Jakob Elle %B Proceedings of the 7th Northern Lights Deep Learning Conference (NLDL) %C Proceedings of Machine Learning Research %D 2026 %E Hyeongji Kim %E Adín Ramírez Rivera %E Benjamin Ricaud %F pmlr-v307-kupcikevicius26a %I PMLR %P 263--273 %U https://proceedings.mlr.press/v307/kupcikevicius26a.html %V 307 %X Laparoscopic ultrasound (LUS) is essential for assessing the liver during laparoscopic liver resections. However, the interpretation of LUS images presents significant challenges due to the steep learning curve and image noise. In this study, we propose an enhanced U-Net-based neural network with a ResNet18 backbone specifically designed for real-time liver vessel segmentation of 2D LUS images. Our approach incorporates five preprocessing steps aimed at maximizing the training information extracted from the ultrasound sonogram region. The modified U-Net model achieved a Dice coefficient of 0.879, demonstrating real-time performance at 40 frames per second and enabling the development of advanced ultrasound-based surgical navigation solutions.
APA
Kupcikevicius, I., Boretto, L., Grunbeck, I.A., Kumar, R.P., Nainamalai, V., Akhavi, M.S., Edwin, B. & Elle, O.J.. (2026). AI-Enabled Vessels Segmentation Model for Real-Time Laparoscopic Ultrasound Imaging. Proceedings of the 7th Northern Lights Deep Learning Conference (NLDL), in Proceedings of Machine Learning Research 307:263-273 Available from https://proceedings.mlr.press/v307/kupcikevicius26a.html.

Related Material