Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 May 15;15(10):1254.
doi: 10.3390/diagnostics15101254.

Object Detection in Laparoscopic Surgery: A Comparative Study of Deep Learning Models on a Custom Endometriosis Dataset

Andrey Bondarenko  1 , Vilen Jumutc  1 , Antoine Netter  2 , Fanny Duchateau  2 , Henrique Mendonca Abrão  3 , Saman Noorzadeh  4 , Giuseppe Giacomello  4   5 , Filippo Ferrari  4   6 , Nicolas Bourdel  4   7 , Ulrik Bak Kirk  8   9 , Dmitrijs Bļizņuks  1
Affiliations

Object Detection in Laparoscopic Surgery: A Comparative Study of Deep Learning Models on a Custom Endometriosis Dataset

Andrey Bondarenko et al. Diagnostics (Basel). .

Abstract

Background: Laparoscopic surgery for endometriosis presents unique challenges due to the complexity of and variability in lesion appearances within the abdominal cavity. This study investigates the application of deep learning models for object detection in laparoscopic videos, aiming to assist surgeons in accurately identifying and localizing endometriosis lesions and related anatomical structures. A custom dataset was curated, comprising of 199 video sequences and 205,725 frames. Of these, 17,560 frames were meticulously annotated by medical professionals. The dataset includes object detection annotations for 10 object classes relevant to endometriosis, alongside segmentation masks for some classes. Methods: To address the object detection task, we evaluated the performance of two deep learning models-FasterRCNN and YOLOv9-under both stratified and non-stratified training scenarios. Results: The experimental results demonstrated that stratified training significantly reduced the risk of data leakage and improved model generalization. The best-performing FasterRCNN object detection model achieved a high average test precision of 0.9811 ± 0.0084, recall of 0.7083 ± 0.0807, and mAP50 (mean average precision at 50% overlap) of 0.8185 ± 0.0562 across all presented classes. Despite these successes, the study also highlights the challenges posed by the weak annotations and class imbalances in the dataset, which impacted overall model performances. Conclusions: In conclusion, this study provides valuable insights into the application of deep learning for enhancing laparoscopic surgical precision in endometriosis treatment. The findings underscore the importance of robust dataset curation and advanced training strategies in developing reliable AI-assisted tools for surgical interventions. The latter could potentially improve the guidance of surgical interventions and prevent blind spots occurring in difficult to reach abdominal regions. Future work will focus on refining the dataset and exploring more sophisticated model architectures to further improve detection accuracy.

Keywords: RCNN; deep learning; endometriosis; object detection.

PubMed Disclaimer

Conflict of interest statement

Authors: Saman Noorzadeh, Giuseppe Giacomello, Filippo Ferrari and Nicolas Bourdel were employed by the company SurgAR. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Object detection distribution of the number of unique annotated objects per video.
Figure 2
Figure 2
Object detection distribution of video lengths (in number of the frames).
Figure 3
Figure 3
YOLOv9 stratified train/validation/test precision, recall, and F-1 curves.
Figure 4
Figure 4
YOLOv9 non-stratified train/validation/test precision, recall, and F-1 curves.
Figure 5
Figure 5
FasterRCNN stratified train/validation/test precision, recall, and F-1 curves.
Figure 6
Figure 6
FasterRCNN non-stratified train/validation/test precision, recall, and F-1 curves.
Figure 7
Figure 7
Test performance metrics by model and split.
Figure 8
Figure 8
FasterRCNN train (left) and validation (right) loss convergences at the stratified split.
Figure 9
Figure 9
YOLOv9 ground-truth and predicted bounding boxes with the corresponding classes.
Figure 10
Figure 10
FasterRCNN ground-truth and predicted bounding boxes with the corresponding classes.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Hummelshoj L., Prentice A., Groothuis P. Update on Endometriosis: 9th World Congress on Endometriosis, 14–17 September 2005, Maastricht, the Netherlands. Women’s Health. 2006;2:53–56. doi: 10.2217/17455057.2.1.53. - DOI - PubMed
    1. Beata S., Szyłło K., Romanowicz H. Endometriosis: Epidemiology, Classification, Pathogenesis, Treatment and Genetics (Review of Literature) Int. J. Mol. Sci. 2021;22:10554. doi: 10.3390/ijms221910554. - DOI - PMC - PubMed
    1. Vobugari N., Raja V., Sethi U., Gandhi K., Raja K., Surani S.R. Advancements in oncology with artificial intelligence—A review article. Cancers. 2022;14:1349. doi: 10.3390/cancers14051349. - DOI - PMC - PubMed
    1. Quazi S. Artificial intelligence and machine learning in precision and genomic medicine. Med. Oncol. 2022;39:120. doi: 10.1007/s12032-022-01711-1. - DOI - PMC - PubMed
    1. Ghaderzadeh M., Shalchian A., Irajian G., Sadeghsalehi H., Zahedi bialvaei A., Sabet B. Artificial Intelligence in Drug Discovery and Development Against Antimicrobial Resistance: A Narrative Review. Iran. J. Med. Microbiol. 2024;18:135–147. doi: 10.30699/ijmm.18.3.135. - DOI

LinkOut - more resources