1
|
Yoon MS, Kwon G, Oh J, Ryu J, Lim J, Kang BK, Lee J, Han DK. Effect of Contrast Level and Image Format on a Deep Learning Algorithm for the Detection of Pneumothorax with Chest Radiography. J Digit Imaging 2023; 36:1237-1247. [PMID: 36698035 PMCID: PMC10287877 DOI: 10.1007/s10278-022-00772-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2022] [Revised: 12/23/2022] [Accepted: 12/29/2022] [Indexed: 01/26/2023] Open
Abstract
Under the black-box nature in the deep learning model, it is uncertain how the change in contrast level and format affects the performance. We aimed to investigate the effect of contrast level and image format on the effectiveness of deep learning for diagnosing pneumothorax on chest radiographs. We collected 3316 images (1016 pneumothorax and 2300 normal images), and all images were set to the standard contrast level (100%) and stored in the Digital Imaging and Communication in Medicine and Joint Photographic Experts Group (JPEG) formats. Data were randomly separated into 80% of training and 20% of test sets, and the contrast of images in the test set was changed to 5 levels (50%, 75%, 100%, 125%, and 150%). We trained the model to detect pneumothorax using ResNet-50 with 100% level images and tested with 5-level images in the two formats. While comparing the overall performance between each contrast level in the two formats, the area under the receiver-operating characteristic curve (AUC) was significantly different (all p < 0.001) except between 125 and 150% in JPEG format (p = 0.382). When comparing the two formats at same contrast levels, AUC was significantly different (all p < 0.001) except 50% and 100% (p = 0.079 and p = 0.082, respectively). The contrast level and format of medical images could influence the performance of the deep learning model. It is required to train with various contrast levels and formats of image, and further image processing for improvement and maintenance of the performance.
Collapse
Affiliation(s)
- Myeong Seong Yoon
- Department of Emergency Medicine, College of Medicine, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
- Machine Learning Research Center for Medical Data, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
- Department of Radiological Science, Eulji University, 553 Sanseong-daero, Seongnam-si, Gyeonggi Do, 13135, Republic of Korea
| | - Gitaek Kwon
- Department of Computer Science, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
- VUNO, Inc, 479 Gangnam-daero, Seocho-gu, Seoul, 06541, Republic of Korea
| | - Jaehoon Oh
- Department of Emergency Medicine, College of Medicine, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea.
- Machine Learning Research Center for Medical Data, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea.
| | - Jongbin Ryu
- Department of Software and Computer Engineering, Ajou University, 206 World cup-ro, Suwon-si, Gyeonggi Do, 16499, Republic of Korea.
| | - Jongwoo Lim
- Department of Computer Science, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
- Machine Learning Research Center for Medical Data, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
| | - Bo-Kyeong Kang
- Machine Learning Research Center for Medical Data, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
- Department of Radiology, College of Medicine, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
| | - Juncheol Lee
- Department of Emergency Medicine, College of Medicine, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul, 04763, Republic of Korea
| | - Dong-Kyoon Han
- Department of Radiological Science, Eulji University, 553 Sanseong-daero, Seongnam-si, Gyeonggi Do, 13135, Republic of Korea
| |
Collapse
|