Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Nov 14;12(11):23259671241291920.
doi: 10.1177/23259671241291920. eCollection 2024 Nov.

A Novel Machine Learning Model to Predict Revision ACL Reconstruction Failure in the MARS Cohort

MARS GroupKinjal Vasavada  1 Vrinda Vasavada  2 Jay Moran  1 Sai Devana  3 Changhee Lee  4 Sharon L Hame  3 Laith M Jazrawi  2 Orrin H Sherman  2 Laura J Huston  5 Amanda K Haas  6 Christina R Allen  1 Daniel E Cooper  7 Thomas M DeBerardino  8 Kurt P Spindler  9 Michael J Stuart  10 Annunziato Ned Amendola  11 Christopher C Annunziata  12 Robert A Arciero  13 Bernard R Bach Jr  14 Champ L Baker 3rd  15 Arthur R Bartolozzi  16 Keith M Baumgarten  17 Jeffrey H Berg  18 Geoffrey A Bernas  19 Stephen F Brockmeier  20 Robert H Brophy  6 Charles A Bush-Joseph  14 J Brad Butler V  21 James L Carey  22 James E Carpenter  23 Brian J Cole  14 Jonathan M Cooper  24 Charles L Cox  5 R Alexander Creighton  25 Tal S David  26 Warren R Dunn  27 David C Flanigan  28 Robert W Frederick  29 Theodore J Ganley  30 Charles J Gatt Jr  31 Steven R Gecha  32 James Robert Giffin  33 Jo A Hannafin  34 Norman Lindsay Harris Jr  35 Keith S Hechtman  36 Elliott B Hershman  37 Rudolf G Hoellrich  38 David C Johnson  39 Timothy S Johnson  39 Morgan H Jones  40 Christopher C Kaeding  28 Ganesh V Kamath  25 Thomas E Klootwyk  41 Bruce A Levy  42 C Benjamin Ma  43 G Peter Maiers 2nd  41 Robert G Marx  34 Matthew J Matava  6 Gregory M Mathien  44 David R McAllister  45 Eric C McCarty  46 Robert G McCormack  47 Bruce S Miller  23 Carl W Nissen  48 Daniel F O'Neill  49 Brett D Owens  50 Richard D Parker  9 Mark L Purnell  51 Arun J Ramappa  52 Michael A Rauh  19 Arthur C Rettig  41 Jon K Sekiya  23 Kevin G Shea  53 James R Slauterbeck  54 Matthew V Smith  6 Jeffrey T Spang  25 Steven J Svoboda  55 Timothy N Taft  25 Joachim J Tenuta  56 Edwin M Tingstad  57 Armando F Vidal  58 Darius G Viskontas  59 Richard A White  60 James S Williams Jr  61 Michelle L Wolcott  62 Brian R Wolf  63 Rick W Wright  5 James J York  64
Affiliations

A Novel Machine Learning Model to Predict Revision ACL Reconstruction Failure in the MARS Cohort

MARS Group et al. Orthop J Sports Med. .

Abstract

Background: As machine learning becomes increasingly utilized in orthopaedic clinical research, the application of machine learning methodology to cohort data from the Multicenter ACL Revision Study (MARS) presents a valuable opportunity to translate data into patient-specific insights.

Purpose: To apply novel machine learning methodology to MARS cohort data to determine a predictive model of revision anterior cruciate ligament reconstruction (rACLR) graft failure and features most predictive of failure.

Study design: Cohort study; Level of evidence, 3.

Methods: The authors prospectively recruited patients undergoing rACLR from the MARS cohort and obtained preoperative radiographs, surgeon-reported intraoperative findings, and 2- and 6-year follow-up data on patient-reported outcomes, additional surgeries, and graft failure. Machine learning models including logistic regression (LR), XGBoost, gradient boosting (GB), random forest (RF), and a validated ensemble algorithm (AutoPrognosis) were built to predict graft failure by 6 years postoperatively. Validated performance metrics and feature importance measures were used to evaluate model performance.

Results: The cohort included 960 patients who completed 6-year follow-up, with 5.7% (n = 55) experiencing graft failure. AutoPrognosis demonstrated the highest discriminative power (model area under the receiver operating characteristic curve: AutoPrognosis, 0.703; RF, 0.618; GB, 0.660; XGBoost, 0.680; LR, 0.592), with well-calibrated scores (model Brier score: AutoPrognosis, 0.053; RF, 0.054; GB, 0.057; XGBoost, 0.058; LR, 0.111). The most important features for AutoPrognosis model performance were prior compromised femoral and tibial tunnels (placement and size) and allograft graft type used in current rACLR.

Conclusion: The present study demonstrated the ability of the novel AutoPrognosis machine learning model to best predict the risk of graft failure in patients undergoing rACLR at 6 years postoperatively with moderate predictive ability. Femoral and tibial tunnel size and position in prior ACLR and allograft use in current rACLR were all risk factors for rACLR failure in the context of the AutoPrognosis model. This study describes a unique model that can be externally validated with larger data sets and contribute toward the creation of a robust rACLR bedside risk calculator in future studies.

Registration: NCT00625885 (ClinicalTrials.gov identifier).

Keywords: ACL revision; femoral tunnel; graft failure; machine learning; tibial tunnel.

PubMed Disclaimer

Conflict of interest statement

One or more of the authors has declared the following potential conflict of interest or source of funding: This study was funded by the National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases (grant 5R01-AR060846). See Supplemental Material for individual disclosures. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

Figures

Figure 1.
Figure 1.
AutoPrognosis design diagram.
Figure 2.
Figure 2.
(A) Receiver operating characteristic curves and (B) precision-recall curves for the AutoPrognosis and logistic regression models.
Figure 3.
Figure 3.
Partial dependence plots of predicted risk of graft failure after revision anterior cruciate ligament reconstruction (ACLR) based on (A) graft type used in current revision ACLR, (B) femoral tunnel placement in prior ACLR, and (C) tibial tunnel placement in prior ACLR.

References

    1. Alaa A, van der Schaar M. AutoPrognosis: automated clinical prognostic modeling via Bayesian optimization with structured kernel learning. Presented at: Proceedings of the 35th International Conference on Machine Learning; July 10-15, 2018; Stockholm, Sweden. Accessed October 13, 2024. https://proceedings.mlr.press/v80/alaa18b.html
    1. Anaconda. Anaconda software distribution. Accessed October 13, 2024. https://anaconda.com
    1. Beam AL, Kohane IS. Big data and machine learning in health care. JAMA. 2018;319(13):1317-1318. - PubMed
    1. Bedi A, Maak T, Musahl V, et al.. Effect of tibial tunnel position on stability of the knee after anterior cruciate ligament reconstruction: is the tibial tunnel position most important? Am J Sports Med. 2011;39(2):366-373. - PubMed
    1. Branche K, Bradsell HL, Lencioni A, Frank RM. Sex-Based Differences in Adult ACL Reconstruction Outcomes. Curr Rev Musculoskelet Med. 2022;15(6):645-650 - PMC - PubMed

Associated data

LinkOut - more resources