Deep-learning-based image identification method of melanoma of skin cancer

Abstract

The invention discloses a deep-learning-based image identification method of the melanoma of a skin cancer. The method comprises: establishing a skin lesion dermatoscope image database; carrying out data preprocessing and quality assessment and screening; carrying out cascaded connection of deep convolution neural networks, and introducing transfer learning and a classifier. At a training stage, enhancement or screening is carried out on original data; and after inputting of positive and negative samples, sample expansion and overfitting prevention are carried out. At the preprocessing stage,data enhancement is added, cascaded connection of two deep convolution neural networks is carried out; transfer learning of existing pre-trained features at a natural image to an identification network is carried out; and then classification prediction is carried out by using the classifier and fine adjustment of network parameters is carried out based on network convergence and prediction situations. Therefore, the accuracy of skin lesion classification is improved; the restriction of manual feature selection is avoided; the adaptability is improved; and thus the deep-learning-based image identification method has the certain significance in a medical skin disease image analysis.

Description

technical field [0001] The present invention relates to the field of artificial intelligence, more specifically, it relates to an image recognition method of skin cancer melanoma based on deep learning. Background technique [0002] Malignant melanoma is a malignant tumor induced by abnormal proliferation of human melanocytes on the surface of the skin, commonly known as skin malignant melanoma. The early symptoms of skin malignant melanoma are not obvious and are easy to be ignored by patients. It is difficult to diagnose early, and the degree of malignancy is extremely high. It is very easy to spread and the prognosis is poor. Usually, the cancer cells have spread to the Certain tissues and organs, even the whole body. Therefore, the early diagnosis and differentiation of malignant melanoma has extremely important social significance and practical value. [0003] Currently, only a few machine learning algorithms are used to identify its signatures to detect melanoma. Th...

AI Technical Summary

Problems solved by technology

However, the accuracy of diagnosing melanoma is still very low if the doctor is inexperienced

[0004] With the advancement and development of technology, in clinical diagnosis, doctors have developed a variety of different diagnostic criteria based on the characteristics of the surface and growth of malignant melanoma. Among them, the widely used diagnostic criteria include pattern analysis, ABCD principles, seven Point inspection method and Montessori method, but these feature extraction methods are relatively complicated, and generally need to be completed manually in actual use, and these manual feature extraction methods can easily lead to the loss of part of the feature information, making diagnosis and identification difficult. The performance is not ideal, and it mainly depends on the experience of the doctor for identification. If the doctor is not experienced enough, there will be misjudgment, so it needs to be further improved

Images