2D segmentation

Methodology:

Meiburger, K. M., Zahnd, G., Faita, F., Loizou, C. P., Carvalho, C., Steinman, D. A., Gibello, L., Bruno, R. M., Marzola, F., Clarenbach, R., Francesconi, M., Nicolaides, A. N., Campilho, A., Ghotbi, R., Kyriacou, E., Navab, N., Griffin, M., Panayiotou, A. G., Gherardini, R., … Molinari, F. (2021). Carotid ultrasound boundary study (Cubs): An open multicenter analysis of Computerized Intima–media thickness measurement systems and their clinical impact. Ultrasound in Medicine & Biology, 47(8), 2442–2455. https://doi.org/10.1016/j.ultrasmedbio.2021.03.022 ) <<< reference of US images and dataset explanation

CIMT: Is calculated from the segmentation output, by calculating the thickness of the segmented part of the artery. It has been calculated how many pixels thick the part is and took the maximum thickness for each image.

Performance metrics: are calculated to evaluate the performance of the segmentation task

IoU:

DSC:

Precision:

Sensitivity:

Specificity:

True positive Rate (TPR):

False Positive Rate (FPR):

Accuracy (ACC):