Journal of Biomedical Engineering Research (대한의용생체공학회:의공학회지)

The Korean Society of Medical and Biological Engineering (대한의용생체공학회)
권호 표지
DOI QR코드

DOI QR Code

A Comparative Performance Analysis of Segmentation Models for Lumbar Key-points Extraction

요추 특징점 추출을 위한 영역 분할 모델의 성능 비교 분석

  • Seunghee Yoo (Digital Health Research Division, Korea Institute of Oriental Medicine) ;
  • Minho Choi (Digital Health Research Division, Korea Institute of Oriental Medicine) ;
  • Jun-Su Jang (Digital Health Research Division, Korea Institute of Oriental Medicine)
  • 유승희 (한국한의학연구원 디지털임상연구부) ;
  • 최민호 (한국한의학연구원 디지털임상연구부) ;
  • 장준수 (한국한의학연구원 디지털임상연구부)
  • Received : 2023.09.26
  • Accepted : 2023.10.27
  • Published : 2023.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Most of spinal diseases are diagnosed based on the subjective judgment of a specialist, so numerous studies have been conducted to find objectivity by automating the diagnosis process using deep learning. In this paper, we propose a method that combines segmentation and feature extraction, which are frequently used techniques for diagnosing spinal diseases. Four models, U-Net, U-Net++, DeepLabv3+, and M-Net were trained and compared using 1000 X-ray images, and key-points were derived using Douglas-Peucker algorithms. For evaluation, Dice Similarity Coefficient(DSC), Intersection over Union(IoU), precision, recall, and area under precision-recall curve evaluation metrics were used and U-Net++ showed the best performance in all metrics with an average DSC of 0.9724. For the average Euclidean distance between estimated key-points and ground truth, U-Net was the best, followed by U-Net++. However the difference in average distance was about 0.1 pixels, which is not significant. The results suggest that it is possible to extract key-points based on segmentation and that it can be used to accurately diagnose various spinal diseases, including spondylolisthesis, with consistent criteria.

Keywords

Acknowledgement

본 연구는 한국한의학연구원의 AI 한의사 개발을 위한 ICT 기반 한의 중점 질환 진단 예측 기술 개발 과제(KSN1823130)의 지원을 받아 수행하였음.

References

  1. Waxenbaum JA, Reddy V, Wiliams C, Futterman B. Anatomy, Back, Lumbar Vertebrae. StatPerals. 2023;29083618.
  2. Nguyen TP, Chae DS, Part SJ, Kang KY, Yoon JH. Deep learning system for Meyerding classification and segmental motion measurement in diagnosis of lumbar spondylolisthesis. Biomedical Signal Processing and Control. 2021;65:102371.
  3. Jang JS. A Quantification method of Lumbar Vertebral Displacement using Polynomial Curves and Convolutional Neural Networks. Journal of Next-generation Convergence Information Services Technology. 2021;10(5):549-57. https://doi.org/10.29056/jncist.2021.10.07
  4. Afif M, Ayachi R, Said Y, Atri M. Deep learning-based technique for lesions segmentation in CT scans images for COVID-19 prediction. Multimedia Tools and Applications. 2023;82:26885-26899. https://doi.org/10.1007/s11042-023-14941-w
  5. Lu HJ, Li MY, Yu K, Zhang YJ, Yu L. Lumbar spine segmentation method based on deep learning. Journal of Applied Clinical Medical Physics. 2023;24(6).
  6. Guinebert S. Petit E, Bousson V, Bodard S, Amoretti N, Kastler B, Automatic semantic segmentation and detection of vertebras and intervertebral discs by neural networks. Computer Methods and Programs in Biomedicine Update. 2022;2:100055.
  7. Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015. Cham: Springer International Publishing; 2015.
  8. Mehta R, Sivaswamy J. M-net: A Convolutional Neural Network for deep brain structure segmentation. 2017 IEEE 14th international symposium on biomedical imaging. 2017;437-40.
  9. Chen LC, Zhu Y, Papandreou G, Schroff F, Adam H. Encdoer-Decoder with Atrous Separabel Convolution for Semantic Image Segmentation. Computer Vision - ECCV 2018. Cham: Springer International Publishing; 2015;833-51.
  10. Zhou Z, Siddiquee MMR, Tajbakhsh N, Liang J. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. DLMIA ML-CDS 2018, Cham: Springer International Publishing; 2018;3-11.
  11. Badrinarayanan V, Kendall A, Cipolla R. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Patter Analysis and Machine Intelligence. 2017;39(12):2481-95. https://doi.org/10.1109/TPAMI.2016.2644615
  12. Klinwichit P, Yookwan W, Limchareon S, Chinnsarn K, Jang JS, Onuean A. BUU-LSPINE: A Thai Open Lumbar Spine Dataset for Spondylolisthesis Detection. Applied Sciences. 2023;13(15):8646.
  13. Kim DH, Jeong JG, Kim YJ, Kim KG, Jeon JY. Automated Vertebral Segmentation and Measurement of Vertebral Compression Ratio Based on Deep Learning in X-Ray Images. J Digit Imaging. 2021;34(4):853-61. https://doi.org/10.1007/s10278-021-00471-0
  14. Baek MH, Jin GJ, Kim YJ, Kim KG, Jeon JY. An Automated Vertebra Segmentation model based on Deep learning and an Application to Cobb angle Measurement based on Spine X-ray. Journal of Next-generation Convergence Information Services Technology. 2020;9(1):1-9. https://doi.org/10.29056/jncist.2020.03.01
  15. Kim KC, Cho HC, Jang Tj, Choi JM, Seo JK. Automatic detection and segmentation of lumbar vertebrae from X-ray images for compression fracture evaluation. Computer Methods and Programs in Biomedicine. 2021;200:105833.
  16. Douglas DH, Peucker TK. Algorithms for the reduction of the number of points required to represent a digitized line or its caricature. Cartographica: The International Journal for Geographic Information and Geovisualization. 1973;10(2):112-122. https://doi.org/10.3138/FM57-6770-U75U-7727
  17. Zuiderveld K. Contrast Limited Adaptive Histogram Equalization. Graphics Gems 4th ed. San Diego: Academic Press Inc; 1994;474-85.
  18. Saenpaen J, Arwatchananukul S, Aunsri N. A Comparison of Image Enhancement Methods for Lumbar Spine X-ray Image. 2018 15th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), 2018;798-801.