Volume 33, Issue 1 p. 59-69
KNEE

Enhanced reliability and time efficiency of deep learning-based posterior tibial slope measurement over manual techniques

Shang-Yu Yao

Shang-Yu Yao

Department of Orthopedic Surgery, Linkou Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Search for more papers by this author
Xue-Zhi Zhang

Xue-Zhi Zhang

Engineering Product Development, Singapore University of Technology and Design, Tampines, Singapore

Search for more papers by this author
Soumyajit Podder

Soumyajit Podder

Department of Biomedical Engineering, Chang Gung University, Taoyuan City, Taiwan

Search for more papers by this author
Chen-Te Wu

Chen-Te Wu

Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Search for more papers by this author
Yi-Shen Chan

Yi-Shen Chan

Department of Orthopedic Surgery, Linkou Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Comprehensive Sports Medicine Center, Taoyuan Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Department of Orthopedic Surgery, Keelung Chang Gung Memorial Hospital, Keelung City, Taiwan

Search for more papers by this author
Dan Berco

Corresponding Author

Dan Berco

Comprehensive Sports Medicine Center, Taoyuan Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Department of Electronics Engineering and Program in Nano-Electronic Engineering and Design, Chang Gung University, Taoyuan City, Taiwan

Correspondence Cheng-Pang Yang, Department of Orthopedic Surgery, Linkou Chang Gung Memorial Hospital, No. 5, Fuxing St, Guishan, Taoyuan City 333, Taiwan.

Email: [email protected]

Dan Berco, Department of Electronics Engineering and Program in Nano-Electronic Engineering and Design, Chang Gung University, No. 259 Wenhua 1st Rd, Guishan, Taoyuan City 333, Taiwan.

Email: [email protected]

Search for more papers by this author
Cheng-Pang Yang

Corresponding Author

Cheng-Pang Yang

Department of Orthopedic Surgery, Linkou Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Comprehensive Sports Medicine Center, Taoyuan Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Correspondence Cheng-Pang Yang, Department of Orthopedic Surgery, Linkou Chang Gung Memorial Hospital, No. 5, Fuxing St, Guishan, Taoyuan City 333, Taiwan.

Email: [email protected]

Dan Berco, Department of Electronics Engineering and Program in Nano-Electronic Engineering and Design, Chang Gung University, No. 259 Wenhua 1st Rd, Guishan, Taoyuan City 333, Taiwan.

Email: [email protected]

Search for more papers by this author
First published: 26 May 2024

Shang-Yu Yao, Xue-Zhi Zhang, and Soumyajit Podder contributed equally to this study and as first authors.

Abstract

Purpose

Multifaceted factors contribute to inferior outcomes following anterior cruciate ligament (ACL) reconstruction surgery. A particular focus is placed on the posterior tibial slope (PTS). This study introduces the integration of machine learning and artificial intelligence (AI) for efficient measurements of tibial slopes on magnetic resonance imaging images as a promising solution. This advancement aims to enhance risk stratification, diagnostic insights, intervention prognosis and surgical planning for ACL injuries.

Methods

Images and demographic information from 120 patients who underwent ACL reconstruction surgery were used for this study. An AI-driven model was developed to measure the posterior lateral tibial slope using the YOLOv8 algorithm. The accuracy of the lateral tibial slope, medial tibial slope and tibial longitudinal axis measurements was assessed, and the results reached high levels of reliability. This study employed machine learning and AI techniques to provide objective, consistent and efficient measurements of tibial slopes on MR images.

Results

Three distinct models were developed to derive AI-based measurements. The study results revealed a substantial correlation between the measurements obtained from the AI models and those obtained by the orthopaedic surgeon across three parameters: lateral tibial slope, medial tibial slope and tibial longitudinal axis. Specifically, the Pearson correlation coefficients were 0.673, 0.850 and 0.839, respectively. The Spearman rank correlation coefficients were 0.736, 0.861 and 0.738, respectively. Additionally, the interclass correlation coefficients were 0.63, 0.84 and 0.84, respectively.

Conclusion

This study establishes that the deep learning-based method for measuring posterior tibial slopes strongly correlates with the evaluations of expert orthopaedic surgeons. The time efficiency and consistency of this technique suggest its utility in clinical practice, promising to enhance workflow, risk assessment and the customization of patient treatment plans.

Level of Evidence

Level III, cross-sectional diagnostic study.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT

All relevant data required for training and testing of the AI model can be found at Open Science Framework (https://osf.io/5tazu/), following an embargo from the date of publication to allow for commercialization of research findings.