abstract details

The summaries are free for public use. ARTHROS will continue to add and archive summaries of articles deemed relevant to ARTHROS by our Faculty.

A neural network to predict the knee adduction moment in patients with osteoarthritis using anatomical landmarks obtainable from 2D video analysis

Author

Osteoarthritis Cartilage. 2021 Jan 7;S1063-4584(21)00003-0. doi: 10.1016/j.joca.2020.12.017.Online ahead of print.

M A Boswell 1S D Uhlrich 2L Kidzinski 3K Thomas 4J A Kolesar 5G E Gold 6G S Beaupre 7S L Delp 8

Author Information

1 Department of Bioengineering, Stanford University, Stanford, CA, USA. Electronic address: boswellm@stanford.edu.

2 Department of Mechanical Engineering, Stanford University, Stanford, CA, USA; Musculoskeletal Research Lab, VA Palo Alto Healthcare System, Palo Alto, CA, USA. Electronic address: suhlrich@stanford.edu.

3 Department of Bioengineering, Stanford University, Stanford, CA, USA. Electronic address: lukasz.kidzinski@stanford.edu.

4 Department of Biomedical Data Science, Stanford University, Stanford, CA, USA. Electronic address: kathoma@stanford.edu.

5 Department of Bioengineering, Stanford University, Stanford, CA, USA; Musculoskeletal Research Lab, VA Palo Alto Healthcare System, Palo Alto, CA, USA. Electronic address: julie14@stanford.edu.

6 Department of Radiology, Stanford University, Stanford, CA, USA. Electronic address: gold@stanford.edu.

7 Department of Bioengineering, Stanford University, Stanford, CA, USA; Musculoskeletal Research Lab, VA Palo Alto Healthcare System, Palo Alto, CA, USA. Electronic address: garybeaupre@gmail.com.

8 Department of Bioengineering, Stanford University, Stanford, CA, USA; Department of Mechanical Engineering, Stanford University, Stanford, CA, USA; Department of Orthopaedic Surgery, Stanford University, Stanford, CA, USA. Electronic address: delp@stanford.edu.

Abstract

Objective: The knee adduction moment (KAM) can inform treatment of medial knee osteoarthritis; however, measuring the KAM requires an expensive gait analysis laboratory. We evaluated the feasibility of predicting the peak KAM during natural and modified walking patterns using the positions of anatomical landmarks that could be identified from video analysis.

Method: Using inverse dynamics, we calculated the KAM for 86 individuals (64 with knee osteoarthritis, 22 without) walking naturally and with foot progression angle modifications. We trained a neural network to predict the peak KAM using the 3-dimensional positions of 13 anatomical landmarks measured with motion capture (3D neural network). We also trained models to predict the peak KAM using 2-dimensional subsets of the dataset to simulate 2-dimensional video analysis (frontal and sagittal plane neural networks). Model performance was evaluated on a held-out, 8-person test set that included steps from all trials.

Results: The 3D neural network predicted the peak KAM for all test steps with r2( Murray et al., 2012) 2 = 0.78. This model predicted individuals' average peak KAM during natural walking with r2( Murray et al., 2012) 2 = 0.86 and classified which 15° foot progression angle modifications reduced the peak KAM with accuracy = 0.85. The frontal plane neural network predicted peak KAM with similar accuracy (r2( Murray et al., 2012) 2 = 0.85) to the 3D neural network, but the sagittal plane neural network did not (r2( Murray et al., 2012) 2 = 0.14).

Conclusion: Using the positions of anatomical landmarks from motion capture, a neural network accurately predicted the peak KAM during natural and modified walking. This study demonstrates the feasibility of measuring the peak KAM using positions obtainable from 2D video analysis.