Dataset for Quantitative Outcome-Oriented Assessment of Microsurgical Anastomosis

This dataset accompanies the paper “Quantitative Outcome-Oriented Assessment of Microsurgical Anastomosis” (EMBC 2025). It provides images and annotations for the instance segmentation of vessels and stitches, supporting image-based assessment of completed microsurgical anastomoses. The dataset contains 50 selected photographs from three recording sessions in Cluj-Napoca, Romania, and Rotterdam, the Netherlands. Vessel and stitch annotations are provided in COCO JSON format. The deposited files include: EMBC2025_All.zip: the complete set of 50 images and their COCO annotations. EMBC2025_Dataset_5_fold.zip: data organized into five folds, including training and test folders, annotations, and visualization files. epoch=395-train_loss=0.161.ckpt: a trained model checkpoint provided with the dataset. Training, inference, and ALI analysis code, together with installation and usage instructions, are available at:https://github.com/luyinEPFL/EMBC2025_ALI Related publication:Hu, L., Gholami, S., Dindelegan, G., Meling, T. R., and Billard, A. (2025). Quantitative Outcome-Oriented Assessment of Microsurgical Anastomosis. 2025 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 1–7.https://doi.org/10.1109/EMBC58623.2025.11254570