Detection of Tumor in the Liver Using CNN and Mobile Net

The prevalence of liver diseases is increasing in areas with high rates of alcohol addiction and among individuals who consume unhealthy diets. This is becoming a significant problem from both a medical and social perspective. Liver cancer is that type of cancer which grows rapidly and can be fatal. Detecting liver cancer in early stages can decrease mortality rates. The intent of this research is to generate a model that can help healthcare professionals identify the type of tumor in the liver region by examining tissue images from a tumor biopsy. At present, identifying whether a tumor is malicious and need medication requires significant effort, time, and expertise from a pathologist. The proposed model can be used as an initial diagnostic tool by histology specialists. The model will be developed using MobileNet architectures, specifically convolutional neural networks (CNNs), trained on ultrasound images of the liver, which possess the ability to transfer knowledge.