Illustration of the proposed sampling protocol.

Similar Items

• Relevant training splits of the VessMAP dataset. Each row shows the average Dice score obtained on the remaining 96 samples when training a neural network using the four images indicated in the training set column. The standard deviation obtained across five repetitions of the training runs is also shown.
  by: Matheus Viana da Silva (21433572)
  Published: (2025)
• Influence of the training set on the generalizability of a model. For each dataset, models were trained on 100 different training sets containing 4 images each and were validated on the remaining images. The difference between the maximum and minimum Dice scores obtained for each sample across all runs was calculated and averaged over all samples. The standard deviation of the differences is also shown to provide a reference regarding the degree of variation observed among samples.
  by: Matheus Viana da Silva (21433572)
  Published: (2025)
• values for the experiments with different splits generated using the VessMAP metadata. Each experiment name depicts the feature used to split the dataset, followed by (h) if the training set had the largest values of the feature or (l) if the training set had the lowest values. Negative values suggest poor generalization and were marked as bold. *Skel. het: skeleton heterogeneity.
  by: Matheus Viana da Silva (21433572)
  Published: (2025)
• Summary of important blood vessel datasets on the literature.
  by: Matheus Viana da Silva (21433572)
  Published: (2025)
• Example segmentations obtained for sample 6818 of the VessMAP dataset.
  by: Matheus Viana da Silva (21433572)
  Published: (2025)