Evaluation of Deep Learning Model for Scoliosis Pre-Screening Using Preprocessed Chest X-ray Images
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Scoliosis is a three-dimensional deformation of the spine that is a deformity induced by physical or disease- related causes as the spine is rotated abnormally. Early detection has a significant influence on the possibility of non- surgical treatment. To train a deep learning model with preprocessed images and to evaluate the results with and without data augmentation to enable the diagnosis of scoliosis based only on a chest X-ray image. The preprocessed images in which only the spine, rib contours, and some hard tissues were left from the original chest image, were used for learning along with the original images, and three CNN(Convolutional Neural Networks) models (VGG16, ResNet152, and EfficientNet) were selected to proceed with training. The results obtained by training with the pre- processed images showed a superior accuracy to those obtained by training with the original image. When the sco- liosis image was added through data augmentation, the accuracy was further improved, ultimately achieving a classification accuracy of 93.56% with the ResNet152 model using test data. Through supplementation with future research, the method proposed herein is expected to allow the early diagnosis of scoliosis as well as cost reduction by reducing the burden of additional radiographic imaging for disease detection.
Abstract
Scoliosis is a three-dimensional deformation of the spine that is a deformity induced by physical or disease- related causes as the spine is rotated abnormally. Early detection has a significant influence on the possibility of non- surgical treatment. To train a deep learning model with preprocessed images and to evaluate the results with and without data augmentation to enable the diagnosis of scoliosis based only on a chest X-ray image. The preprocessed images in which only the spine, rib contours, and some hard tissues were left from the original chest image, were used for learning along with the original images, and three CNN(Convolutional Neural Networks) models (VGG16, ResNet152, and EfficientNet) were selected to proceed with training. The results obtained by training with the pre- processed images showed a superior accuracy to those obtained by training with the original image. When the sco- liosis image was added through data augmentation, the accuracy was further improved, ultimately achieving a classification accuracy of 93.56% with the ResNet152 model using test data. Through supplementation with future research, the method proposed herein is expected to allow the early diagnosis of scoliosis as well as cost reduction by reducing the burden of additional radiographic imaging for disease detection.
참고문헌 (32)
- Cobb J, "Outline for the study of scoliosis", instructional course lectures, 5(1), pp.261~275 (1948)
- Kadoury S, "Classification of three-dimensional thoracic deformities in adolescent idiopathic scoliosis from a multivariate analysis", Eur. Spine J, 21(1), pp.40~49 (2012)
- Sibel B, "Rule-based fuzzy classifier for spinal deformities", Biomed Mater Eng, 24(6), pp.3311~3119 (2014)
- Imaz CY, "Surgical treatment of adolescent idiopathic scoliosis", M.Sc, pp.2001
- Trobisch P, "Idiopathic Scoliosis", Continuing Medical Education, 107(49), pp.875~884 (2010)
- Cheng JC, "Adolescent idiopathic scoliosis", Nat. Rev. Dis. Prim, 24(1), pp.1~20 (2015)
- Negrini S, "2016 SOSORT guidelines : orthopaedic and rehabilitation treatment of idiopathic scoliosis during growth", Scoliosis, 13(3), pp.1~48 (2018)
- Lenke LG, "Lenke classification system of adolescent idiopathic scoliosis : treatment recommendations", Instr Course Lect, 54, pp.537~542 (2005)
- Sapountzi-Krepia DS, "Perceptions of body image, happiness and satisfaction in adolescents wearing a Boston brace for scoliosis treatment", J Adv Nurs, 35(5), pp.683~690 (2001)
- Reamy BV, "Adolescent idiopathic scoliosis : review and current concepts", Am Fam Physician, 64(1), pp.111~116 (2001)
- Lee MK, "Review on conservative treatment of spinal scoliosis", The Journal of Korea CHUNA Manual Medicine for Spine and Nerves, 4(1), pp.103~117 (2009)
- Lonstein JE, "Voluntary school screening for scoliosis in Minnesota", J Bone Joint Surg Am, 64(4), pp.481~488 (1982)
- Tu Y, "Automatic measurement algorithm of scoliosis Cobb angle based on deep learning", Journal of Physics : Conference Series, 1187(4), pp.042100~ (2019)
- Zhou GQ, "Automatic Measurement of Spine Curvature on 3-D Ultrasound Volume Projection Image with Phase Features", IEEE Transactions on Medical Imaging, 36(6), pp.1250~1262 (2017)
- Sugita K, "Epidemiological study on idiopathic scoliosis in high school students. Prevalence and relation to physique, physical strength and motor ability", Nihon Koshu Eisei Zasshi, 47(4), pp.320~325 (2000)
- 오창현, "Usefulness of Chest Radiographs for Scoliosis Screening: A Comparison with Thoraco-Lumbar Standing Radiographs", Yonsei Medical Journal, 53(6), pp.1183~1189 (2012), 연세대학교의과대학[DOI]
- Vollmer S, "Machine learning and artificial intelligence research for patient benefit: 20 critical questions on transparency, replicability, ethics, and effectiveness", BMJ, 368, pp.l6927~ (2020)
- Gstoettner M, "Inter-and intraobserver reliability assessment of the Cobb angle : manual versus digital measurement tools", Eur Spine J, 16(10), pp.1587~1592 (2007)
- Yang J, "Development and validation of deep learning algorithms for scoliosis screening using back images", Commun Biol, 25(2), pp.390~ (2019)
- Wang H, "Application of deep learning upon spinal radiographs to predict progression in adolescent idiopathic scoliosis at first clinic visit", EClinicalMedicine, 29(42), pp.101220~ (2021)
- Wang J, "A review of deep learning on medical image analysis", International Journal of Multimedia Information Retrieval, 26(2), pp.19~38 (2021)
- Wua H, "Automated comprehensive adolescent idiopathic scoliosis assessment using MVCNet", Med Image Anal, 48, pp.1~11 (2018)
- Song TR, "The research of X-ray bone fracture image enhancement algorithms", International Conference on Computer, Mechatronics, Control and Electronic Engineering, pp.24~26 (2010)
- Calderon S, "Assessing the impact of the deceived non local means filter as a preprocessing stage in a convolutional neural network based approach for age estimation using digital hand X-Ray images", IEEE International Conference on Image Processing (ICIP), pp.7~10 (2018)
- Sreenivasulu N, "Color image enhancement using adaptive sigmoid function with bi-histogram equalization", NCICCT, 3(12), pp.1~7 (2015)
- Shorten C, "A survey on image data augmentation for deep learning", J Big Data, 6(60), pp.1~48 (2019)
- Gu S, "Improve Image Classification Using Data Augmentation and Neural Networks", SMU Data Science Review, 2(2), pp.1~44 (2019)
- Simonyan K, "Very deep convolutional networks for large-scale image recognition", ICLR (2015)
- He K, "Deep residual learning for image recognition", Computer Vision and Pattern Recognition (2015)
- Tan M, "EfficientNet : Rethinking model scaling for convolutional neural networks" (2019), ICML
- Zhang H, "Overfitting and underfitting analysis for deep learning based end-to-end communication systems", International Conference on Wireless Communications and Signal Processing (WCSP), pp.23~25 (2019)
- Nguyen QH, "Influence of data splitting on performance of machine learning models in prediction of shear strength of soil", Mathematical Problems in Engineering, 2021(6), pp.1~15 (2021)