[서울대 학사논문+PPT] 척추 x레이 이미지분석 의료인공지능 모델

소개글

척추 x레이 이미지를 머신러닝으로 분석해서 자동으로 척추의 휜 각도를 추출해내는 모델입니다.

서울대학교 의과대학 연구실에서 연구 후, 졸업논문으로 제출한 내용입니다.

논문 + 발표 ppt첨부입니다.

목차

1. 논문
2. 발표 ppt

본문내용

Abstract
This research suggests the way to automatically calculate Cobb angles from spine X-ray image. Convolutional neural network predicts landmarks, the coordinates of each vertebrae in the image. Spinal curve is extracted by the predicted landmarks by a mathematical algorithm. The network was trained with 609 publicly accessible spine x-ray images. This research achieved SMAPE score of 21.70 on test set, 19.58 on test set with appropriate cropping.

Keywords: deep learning, convolutional neural network, AASCE Challenge, scoliosis, medical imaging, spine, Cobb angle

1. Introduction
Adolescent Idiopathic Scoliosis (AIS) is a condition appearing in late childhood or adolescence, characterized by an abnormal curvature of spine. Cobb angle is clinical standard for diagnosis and treatment of AIS, so accurate prediction of Cobb angle is critical. However, manual Cobb angle measurement is time-consuming and prone to errors. Hence, Automatic Cobb angle estimation has been considerably called for to overcome deficiencies of traditional method. [2]

참고 자료

Dubost F., Collery B., Renaudier A., and Roc A. 2020. ‘Automated Estimation of the Spinal Curvature via Spine Centerline Extraction with Ensembles of Cascaded Neural Networks’, Computational Methods and Clinical Applications for Spine Imaging, 88-94
Wu H., Bailey C., Rasoulinejad P., and Li S. 2017. ‘Automatic landmark estimation for adolescent idiopathic scoliosis assessment using boostnet’, Medical Image Computing and Computer Assisted Intervention – MICCAI 2017, 127-135.
Khanal B., Dahal L., Adhikari P., Khanal B. 2020. ‘Automatic Cobb Angle Detection Using Vertebra Detector and Vertebra Corners Regression’, Computational Methods and Clinical Applications for Spine Imaging, 81-87
Girshick R. 2015. ‘Fast R-CNN’, 2015 IEEE International Conference on Computer Vision (ICCV), 1440-1448.
Kingma D. P., and Ba J. L. 2015. ‘Adam: A method for stochastic optimization’, arXiv preprint, arXiv: 1412.6980
Ioffe S., and Szegedy C. 2015. ‘Batch normalization: Accelerating deep network training by reducing internal covariance shift’, arXiv preprint, arXiv:1502.03167
Srivastava N., Hinton G., Krizhevsky A., Sutskever I., and Salakhutdinov R. 2014. ‘Dropout: A Simple Way to Prevent Neural Networks from Overfitting’, Journal of Machine Learning Research, 15:1929-1958
He K., Zhang X., Ren S. and Sun J. 2016. ‘Deep Residual Learning for Image Recognition’, 2016 IEEE Conference on Computer Vision and Pattern Recognition, 770-778.
‘aasce19-home’, AASCE 2019. 2019, accessed Mar 22 2020, https://aasce19.grand-challenge.org.

압축파일 내 파일목록

Lab_Final results and Contents of Paper.pptx
논문 총합 032223.docx