System and method for estimating the metastatic status of a sentinel lymph node

A computer-implemented method at a processor for estimating a metastatic status of a sentinel lymph node based on an ultrasound image of a primary tumor, the method comprising:

• segmenting said ultrasound image, wherein segmenting said ultrasound image includes determining a region of interest, ROI, comprising an intratumoral region of the primary tumor and a peritumoral region of the primary tumor, and wherein segmenting said ultrasound image comprises:
• performing a semantic segmentation process using a first convolutional neural network, CNN, that has been trained to determine said intratumoral region of the primary tumor;
• and
• extracting a plurality of further features from said entire ultrasound image by means of and within layers of said first CNN, said further features being high-level features, preferably wherein said further features comprise at least one of a shape and an object;
• extracting a plurality of characteristics of said image from said ROI using a transfer learning approach employing a pre-trained CNN that is different than said first CNN;
• aggregating into a vector said plurality of characteristics extracted from said ROI and said further plurality of features extracted from said entire ultrasound image;
• selecting a plurality of characteristics stored in said vector, wherein said plurality of selected features comprises one or more characteristics selected from said plurality of characteristics extracted from said ROI and one or more features selected from said further plurality of features extracted from said entire ultrasound image;
• and
• classifying the metastatic status of said sentinel lymph node based on said plurality of characteristic selected from said vector.