YOSHIMITSU Kitaro    Department   Graduate School of Medical Science, Graduate School of Medical Science    Position   Assistant Professor
Language	English
Title	Classification of Oral Cancer and Leukoplakia Using Oral Images and Deep Learning with Multi-Scale Random Crop Self-Training
Conference	14th NameInternational Conference on Pattern Recognition Applications and Methods
Conference Type	International society and overseas society
Presentation Type	Speech
Lecture Type	General
Publisher and common publisher	◎HAMADA Itsuki, OHKAWAUCHI Takaaki, AOSHIMA Chisa, YOSHIMITSU Kitaro, KAIBUCHI Nobuyuki, OKAMOTO Toshihiro, SAKAGUCHI Katsuhisa, OHYA Jun
Date	2025/02/23
Venue (city and name of the country)	Portugal
Holding period	2025/02/23～2025/02/25
Society abstract	Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods 780-787 2025
Summary	This paper proposes Multi-Scale Random Crop Self-Training (MSRCST) for classifying oral cancers and leukoplakia using oral images acquired by our dermoscope. MSRCST comprises the following three key modules: (1) Multi-Scale Random Crop, which extracts image patches at various scales from high-resolution images, preserving both local details and global contextual information essential for accurate classification, (2) Selection based on Confidence, which employs a teacher model to assign confidence scores to each cropped patch, selecting only those with high confidence for further training and ensuring the model focusing on diagnostically relevant features, (3) Iteration of Self-training, which iteratively retrains the model using the selected high-confidence, pseudo-labeled data, progressively enhancing accuracy. In our experiments, we applied MSRCST to classify images of oral cancer and leukoplakia. When combined with MixUp data augmentation, MSRCST achieved an average classification accuracy of 71.71%, outperforming traditional resizing and random cropping methods.