Kwon, Jangho Choi, Kihwan 2024-01-12T03:44:03Z 2024-01-12T03:44:03Z 2022-04-30 2021-11 1557-170X https://pubs.kist.re.kr/handle/201004/77299 We consider the problem of training a convolutional neural network for histological localization of colorectal lesions from imperfectly annotated datasets. Given that we have a colonoscopic image dataset for 4-class histology classification and another dataset originally dedicated to polyp segmentation, we propose a weakly supervised learning approach to histological localization by training with the two different types of datasets. With the classification dataset, we first train a convolutional neural network to classify colonoscopic images into 4 different histology categories. By interpreting the trained classifier, we can extract an attention map corresponding to the predicted class for each colonoscopy image. We further improve the localization accuracy of attention maps by training the model to focus on lesions under the guidance of the polyp segmentation dataset. The experimental results show that the proposed approach simultaneously improves histology classification and lesion localization accuracy. English IEEE Weakly Supervised Attention Map Training for Histological Localization of Colonoscopy Images Conference 10.1109/EMBC46164.2021.9629608 1 43rd Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (IEEE EMBC), pp.3725 - 3728 43rd Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (IEEE EMBC) 3725 3728 US ELECTR NETWORK 2021-11-01 2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC) 000760910503156 2-s2.0-85122528419