Close
  Indian J Med Microbiol
 

Figure 2: The web application with Tissue Adequacy model prediction scores for different Breast Cancer TMA core images. The “My Box” tab is the file browser that allows the user to navigate through their Box account within the application. The “Annotations” tab lets them define classifications and add labels or comments to an image; the model prediction score is shown immediately adjacent to where the user would assign the label. The thumbs-up, crossed fingers, and thumbs down buttons denote the Optimal, Suboptimal, and Unsatisfactory labels, respectively, and are what the pathologists used to classify the image in one of the three classes. The model score is the activation value of the final layer of the trained neural network and is a continuous value normalized between 0 and 1 – the higher the score, the greater the likelihood of the prediction being true. (a) An example of a core with optimal Tissue Adequacy, with a classification score of 0.938 to the Optimal class by the model. (b) A core showing inadequate tissue, and the model agrees with high “confidence” as can be seen from the classification score of 1 for the Unsatisfactory class

Figure 2: <b>The web application with Tissue Adequacy model prediction scores for different Breast Cancer TMA core images</b>. The “My Box” tab is the file browser that allows the user to navigate through their Box account within the application. The “Annotations” tab lets them define classifications and add labels or comments to an image; the model prediction score is shown immediately adjacent to where the user would assign the label. The thumbs-up, crossed fingers, and thumbs down buttons denote the Optimal, Suboptimal, and Unsatisfactory labels, respectively, and are what the pathologists used to classify the image in one of the three classes. The model score is the activation value of the final layer of the trained neural network and is a continuous value normalized between 0 and 1 – the higher the score, the greater the likelihood of the prediction being true. (a) An example of a core with optimal Tissue Adequacy, with a classification score of 0.938 to the Optimal class by the model. (b) A core showing inadequate tissue, and the model agrees with high “confidence” as can be seen from the classification score of 1 for the Unsatisfactory class