Deep learning approach to classification of lung cytological images: Two-step training using actual and synthesized images by progressive growing of generative adversarial networks

• Published in: PloS one Vol. 15; no. 3; p. e0229951
• Main Authors: Teramoto, Atsushi, Tsukamoto, Tetsuya, Yamada, Ayumi, Kiriyama, Yuka, Imaizumi, Kazuyoshi, Saito, Kuniaki, Fujita, Hiroshi
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 05.03.2020; Public Library of Science (PLoS)
• Subjects: Accuracy; Artificial neural networks; Automation; Benign; Biology and Life Sciences; Bone marrow; Cable television broadcasting industry; Cancer diagnosis; Cancer therapies; Cellular biology; Classification; Computer and Information Sciences; Cytology; Data acquisition; Deep learning; Generative adversarial networks; Image acquisition; Image classification; Image quality; Image resolution; Lung cancer; Lung diseases; Machine learning; Medical diagnosis; Medical imaging; Medical imaging equipment; Medicine; Medicine and Health Sciences; Microscopy; Neural networks; Performance evaluation; Research and Analysis Methods; Synthesis; Training; United States; Japan

Cytology is the first pathological examination performed in the diagnosis of lung cancer. In our previous study, we introduced a deep convolutional neural network (DCNN) to automatically classify cytological images as images with benign or malignant features and achieved an accuracy of 81.0%. To fur...