Computer-aided diagnosis of digestive tract tumor based on deep learning for medical images

With the continuous development of society, natural pollution and people's unhealthy habits have led to an increasing number of patients with gastrointestinal cancer. As a malignant tumor, if the digestive tract tumor can be extracted and checked out, it will be very helpful to the patient's treatment. But the detection of gastrointestinal tumors is really not easy, so this article hopes that the method based on deep learning artificial intelligence will help the key technology of computer-aided diagnosis of gastrointestinal tumors in medical images. Through research, it is found that as the learning rate alpha increases, the running time of the network will decrease. When the network is trained to 700 times, it will converge. When the learning rate alpha is 1.1, the network has the highest recognition accuracy and the shortest running time. When alpha =1.1, after the network iteration 700, the accuracy of the network is very high, so we can think that this article is aimed at the CNN classification model of tumor cell image recognition. After the CNN model is improved and optimized through pre-training and dropout technology, the CNN model can solve the classification problem of tumor cell images very well.