Clinical Decision System to Identify Histogenesis in cases of Carcinoma with Unknown Primary
Machine Learning Model !
The clinical decision support system also utilizes power of automated digital image processing and artificial intelligence to classify histology images for their potential primary group. Three machine learning algorithms viz. support vector machine, random forest and artificial neural network have been applied on the images populated in the knowledgebase to develop predictive models.
Carcinoma histology images of known primaries are used as training set and the model is validated with internal and external test sets.