Identification and Classification of Cancer Cells Using Capsule Network With Pathological Images

Cancer is a deadly disease that is costing the lives of many people. Over 9.6 million death is reported in 2018 due to cancer. We propose an ideal methodology to identify and classify cancer cells using pathological images with the help of capsule network. Capsule network's capability to learn patterns based on previous iterations can be exploited for this purpose. This can help in identification of cancer at early stages and work at the root cause of the disease and walk towards completely shutting down the disease. Image processing is done along with fuzzification and further, it is handled with capsule network classifier and analysed.