InceptionV3 기반의 심장비대증 분류 정확도 향상 연구
A Study on the Improvement of Accuracy of Cardiomegaly Classification Based on InceptionV3
3
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The purpose of this study is to improve the classification accuracy compared to the existing InceptionV3 model by proposing a new model modified with the fully connected hierarchical structure of InceptionV3, which showed excellent performance in medical image classification. The data used for model training were trained after data augmentation on a total of 1026 chest X-ray images of patients diagnosed with normal heart and Cardiomegaly at Kyungpook National University Hospital. As a result of the experiment, the learning classification accuracy and loss of the InceptionV3 model were 99.57% and 1.42, and the accuracy and loss of the proposed model were 99.81% and 0.92. As a result of the classification performance evaluation for precision, recall, and F1 score of Inception V3, the precision of the normal heart was 78%, the recall rate was 100%, and the F1 score was 88. The classification accuracy for Cardiomegaly was 100%, the recall rate was 78%, and the F1 score was 88. On the other hand, in the case of the proposed model, the accuracy for a normal heart was 100%, the recall rate was 92%, and the F1 score was 96. The classification accuracy for Cardiomegaly was 95%, the recall rate was 100%, and the F1 score was 97. If the chest X-ray image for normal heart and Cardiomegaly can be classified using the model proposed based on the study results, better classification will be possible and the reliability of classification performance will gradually increase.
Abstract
The purpose of this study is to improve the classification accuracy compared to the existing InceptionV3 model by proposing a new model modified with the fully connected hierarchical structure of InceptionV3, which showed excellent performance in medical image classification. The data used for model training were trained after data augmentation on a total of 1026 chest X-ray images of patients diagnosed with normal heart and Cardiomegaly at Kyungpook National University Hospital. As a result of the experiment, the learning classification accuracy and loss of the InceptionV3 model were 99.57% and 1.42, and the accuracy and loss of the proposed model were 99.81% and 0.92. As a result of the classification performance evaluation for precision, recall, and F1 score of Inception V3, the precision of the normal heart was 78%, the recall rate was 100%, and the F1 score was 88. The classification accuracy for Cardiomegaly was 100%, the recall rate was 78%, and the F1 score was 88. On the other hand, in the case of the proposed model, the accuracy for a normal heart was 100%, the recall rate was 92%, and the F1 score was 96. The classification accuracy for Cardiomegaly was 95%, the recall rate was 100%, and the F1 score was 97. If the chest X-ray image for normal heart and Cardiomegaly can be classified using the model proposed based on the study results, better classification will be possible and the reliability of classification performance will gradually increase.
참고문헌 (21)
- Philbin EF, "The relationship between cardiothoracic ratio and left ventricular ejection fraction in congestive heart failure", Archives of internal medicine, 158(5), pp.501~506 (1998)
- Anderson JC, "Inability to show clot: one limitation of ultrasonography of the abdominal aorta", Radiol-ogy, 132(3), pp.693~696 (1979)
- 문희정, "갑상선 초음파: 교육에 따른 검사자간 일치도와 진단 정확도의 변화", ULTRASONOGRAPHY, 30(1), pp.23~28 (2011), 대한초음파의학회
- Frohlich ED, "Left ventricular hypertrophy as a risk factor", Cardiology clinics, 4(1), pp.137~144 (1986)
- Levy D, "Echocardiographically detected left ven-tricular hypertrophy: prevalence and risk factors: the Framingham Heart Study", Annals of internal medicine, 108(1), pp.7~13 (1988)
- 임상헌, "딥 러닝 기반의 악성흑색종 분류를 위한 컴퓨터 보조진단 알고리즘", (사)디지털산업정보학회 논문지, 14(4), pp.69~77 (2018), (사)디지털산업정보학회[DOI]
- Wang X, "ChestX-ray8: Hospital-scale chest X-ray database and bench-marks on weakly-supervised classification and localization of common thorax diseases", Proceedings of the IEEE con-ference on computer vision and pattern recognition, pp.2097~2106 (2017)
- 김지율, "Inception V3를 이용한 흉부촬영 X선 영상의 폐렴 진단 분류", 한국방사선학회논문지, 14(6), pp.773~780 (2020), 한국방사선학회[DOI]
- Szegedy C, "Going deeper with convolutions", Proceed-ings of the IEEE conference on computer vision and pattern recognition, pp.1~9 (2015)
- Lambert J, "Deep learning under privi-leged information using heteroscedastic dropout", Proceedings of the IEEE Conference on Computer Vision and Pattern Recog-nition, pp.8886~8895 (2018)
- 송호준, "컨볼루션 뉴럴 네트워크 기반의 딥러닝을 이용한 흉부 X-ray 영상의 분류 및 정확도 평가", 한국방사선학회논문지, 14(1), pp.39~44 (2020), 한국방사선학회[DOI]
- "Advanced Guide to Inception v3 on Cloud TPU"[URL]
- 이종근, "흉부 디지털 영상의 병변 유무 판단을 위한 딥러닝 모델", 한국정보통신학회논문지, 24(2), pp.212~218 (2020), 한국정보통신학회[DOI]
- 송경두, "딥러닝 알고리즘 개발과정을 통해 본 영상의학분야에서 딥러닝의 최신 경향", 대한영상의학회지, 80(2), pp.202~212 (2019), 대한영상의학회[DOI]
- Philbin EF, "Digbetween cardiothoracic ratio and left ventricular ejection fraction in congestive heart failure", Archives of internal medicine, 158(5), pp.501~506 (1998)
- Divide the gradient by a running average of its recent magnitude, "Tieleman T", coursera, ~ (2017)
- 국내 인공지능 (AI) 의료기기 현황 및 규제 이슈, "정원준", IITP, ~ (2018)
- Japkowicz N, "Evaluating learning algorithms: a classification perspective" (2011), Cambridge University Press