Deep convolutional neural networks as a decision support tool in medical problems – malignant melanoma case study

• Published in: Trends in Advanced Intelligent Control, Optimization and Automation pp. 848 - 856
• Main Authors: Kwasigroch, Arkadiusz, Mikołajczyk, Agnieszka, Grochowski, Michał
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing
• Series: Advances in Intelligent Systems and Computing
• Subjects: deep convolutional neural network; deep learning; image analysis; machine learning; malignant melanoma; transfer learning
• ISBN: 9783319606989; 3319606980
• ISSN: 2194-5357; 2194-5365
• DOI: 10.1007/978-3-319-60699-6_81

The paper presents utilization of one of the latest tool from the group of Machine learning techniques, namely Deep Convolutional Neural Networks (CNN), in process of decision making in selected medical problems. After the survey of the most successful applications of CNN in solving medical problems, the paper focuses on the very difficult problem of automatic analyses of the skin lesions. The authors propose the CNN structure and the way to cope with the insufficient number of learning data. The research was carried out and validated on the data base of over 10000 images. The efficiency of the proposed approach reaches 84%.