Machine Learning for MSK Image Processing & Interpretation
Cem M Deniz1
1Radiology, New York University Langone Health, New York, NY, United States

Synopsis

This educational lecture will provide an overview of the machine learning approaches applied in MR image processing and interpretation for musculoskeletal disorders.

Target Audience

Researchers interested in the applications of machine learning to musculoskeletal disorders.

Abstract

Machine learning (ML) has been used in MSK disorders for processing and interpreting MR images. Recent advancements in a special type of machine learning method, called deep learning, have revolutionized image recognition, speech recognition, and natural language processing1. Convolutional neural networks (CNNs), a particular type of deep learning method, have recently been successfully used in medical research for image segmentation and computer-aided diagnosis2. In contrast to previous approaches of applying ML methods in MSK MR images, which rely on the development of hand-crafted features, deep CNNs learn increasingly complex features from data automatically. This flexibility enables the extensive use of CNN approaches in MSK image processing and interpretation, resulting in state-of-the-art segmentation and classification models3–8. In this talk, I will present how machine learning is currently facilitated in MSK image processing and interpretation. I will explain the current deep learning architectures commonly used in tissue segmentation and the diagnosis of MSK disorders. Lastly, I will briefly introduce the current visualization/interpretation approaches for deep learning models. At the end of the lecture, the audience is expected to be able to choose the most appropriate machine learning method for their specific MSK research problems.

Acknowledgements

No acknowledgement found.

References

  1. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436-444. doi:10.1038/nature14539
  2. Litjens G, Kooi T, Bejnordi BE, et al. A Survey on Deep Learning in Medical Image Analysis. 2017;(1995). http://arxiv.org/abs/1702.05747.
  3. Lundervold AS, Lundervold A. An overview of deep learning in medical imaging focusing on MRI. Z Med Phys. 2019;29(2):102-127. doi:10.1016/j.zemedi.2018.11.002
  4. Kijowski R, Liu F, Caliva F, Pedoia V. Deep Learning for Lesion Detection , Progression , and Prediction of Musculoskeletal Disease. 2019:1-13. doi:10.1002/jmri.27001
  5. Chea P, Mandell JC. Current applications and future directions of deep learning in musculoskeletal radiology. Skeletal Radiol. 2019:183-197. doi:10.1007/s00256-019-03284-z
  6. Hirschmann A, Cyriac J, Stieltjes B, Kober T, Richiardi J, Omoumi P. Artificial Intelligence in Musculoskeletal Imaging: Review of Current Literature, Challenges, and Trends. Semin Musculoskelet Radiol. 2019;23(3):304-311. doi:10.1055/s-0039-1684024
  7. Liu F, Kijowski R. Deep Learning in Musculoskeletal Imaging. Adv Clin Radiol. 2019;1:83-94. doi:10.1016/j.yacr.2019.04.013
  8. Gyftopoulos S, Lin D, Knoll F, Doshi AM, Rodrigues TC, Recht MP. Artificial Intelligence in Musculoskeletal Imaging: Current Status and Future Directions. Am J Roentgenol. 2019;213(3):506-513. doi:10.2214/AJR.19.21117
Proc. Intl. Soc. Mag. Reson. Med. 28 (2020)