Detection of pulmonary nodules in chest radiographs: novel cost function for effective network training with purely synthesized datasets.

Autor: Hanaoka S; Department of Radiology, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan. hanaoka-tky@g.ecc.u-tokyo.ac.jp., Nomura Y; Center for Frontier Medical Engineering, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba, Japan.; Department of Computational Diagnostic Radiology and Preventive Medicine, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan., Yoshikawa T; Department of Computational Diagnostic Radiology and Preventive Medicine, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan., Nakao T; Department of Computational Diagnostic Radiology and Preventive Medicine, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan., Takenaga T; Department of Radiology, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan., Matsuzaki H; Center for Epidemiology and Preventive Medicine, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan., Yamamichi N; Center for Epidemiology and Preventive Medicine, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan., Abe O; Department of Radiology, University of Tokyo Hospital, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan.
Jazyk: angličtina
Zdroj: International journal of computer assisted radiology and surgery [Int J Comput Assist Radiol Surg] 2024 Oct; Vol. 19 (10), pp. 1991-2000. Date of Electronic Publication: 2024 Jul 13.
DOI: 10.1007/s11548-024-03227-7
Abstrakt: Purpose: Many large radiographic datasets of lung nodules are available, but the small and hard-to-detect nodules are rarely validated by computed tomography. Such difficult nodules are crucial for training nodule detection methods. This lack of difficult nodules for training can be addressed by artificial nodule synthesis algorithms, which can create artificially embedded nodules. This study aimed to develop and evaluate a novel cost function for training networks to detect such lesions. Embedding artificial lesions in healthy medical images is effective when positive cases are insufficient for network training. Although this approach provides both positive (lesion-embedded) images and the corresponding negative (lesion-free) images, no known methods effectively use these pairs for training. This paper presents a novel cost function for segmentation-based detection networks when positive-negative pairs are available.
Methods: Based on the classic U-Net, new terms were added to the original Dice loss for reducing false positives and the contrastive learning of diseased regions in the image pairs. The experimental network was trained and evaluated, respectively, on 131,072 fully synthesized pairs of images simulating lung cancer and real chest X-ray images from the Japanese Society of Radiological Technology dataset.
Results: The proposed method outperformed RetinaNet and a single-shot multibox detector. The sensitivities were 0.688 and 0.507 when the number of false positives per image was 0.2, respectively, with and without fine-tuning under the leave-one-case-out setting.
Conclusion: To our knowledge, this is the first study in which a method for detecting pulmonary nodules in chest X-ray images was evaluated on a real clinical dataset after being trained on fully synthesized images. The synthesized dataset is available at https://zenodo.org/records/10648433 .
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz: