Evaluation of a Deep Learning Software for Automated Measurements on Full-leg Standing Radiographs

Overview

Journal Knee Surg Relat Res

Publisher Biomed Central

Date 2024 Nov 29

PMID 39614404

Authors

Louis Lassalle

Nor-Eddine Regnard

Marion Durteste

Jeanne Ventre

Vincent Marty

Lauryane Clovis

Zekun Zhang

Nicolas Nitche

Alexis Ducarouge

Jean-Denis Laredo

Ali Guermazi

Affiliations

Soon will be listed here.

Abstract

Background: Precise lower limb measurements are crucial for assessing musculoskeletal health; fully automated solutions have the potential to enhance standardization and reproducibility of these measurements. This study compared the measurements performed by BoneMetrics (Gleamer, Paris, France), a commercial artificial intelligence (AI)-based software, to expert manual measurements on anteroposterior full-leg standing radiographs.

Methods: A retrospective analysis was conducted on a dataset comprising consecutive anteroposterior full-leg standing radiographs obtained from four imaging institutions. Key anatomical landmarks to define the hip-knee-ankle angle, pelvic obliquity, leg length, femoral length, and tibial length were annotated independently by two expert musculoskeletal radiologists and served as the ground truth. The performance of the AI was compared against these reference measurements using the mean absolute error, Bland-Altman analyses, and intraclass correlation coefficients.

Results: A total of 175 anteroposterior full-leg standing radiographs from 167 patients were included in the final dataset (mean age = 49.9 ± 23.6 years old; 103 women and 64 men). Mean absolute error values were 0.30° (95% confidence interval [CI] [0.28, 0.32]) for the hip-knee-ankle angle, 0.75 mm (95% CI [0.60, 0.88]) for pelvic obliquity, 1.03 mm (95% CI [0.91,1.14]) for leg length from the top of the femoral head, 1.45 mm (95% CI [1.33, 1.60]) for leg length from the center of the femoral head, 0.95 mm (95% CI [0.85, 1.04]) for femoral length from the top of the femoral head, 1.23 mm (95% CI [1.12, 1.32]) for femoral length from the center of the femoral head, and 1.38 mm (95% CI [1.21, 1.52]) for tibial length. The Bland-Altman analyses revealed no systematic bias across all measurements. Additionally, the software exhibited excellent agreement with the gold-standard measurements with intraclass correlation coefficient (ICC) values above 0.97 for all parameters.

Conclusions: Automated measurements on anteroposterior full-leg standing radiographs offer a reliable alternative to manual assessments. The use of AI in musculoskeletal radiology has the potential to support physicians in their daily practice without compromising patient care standards.

References

Lu M, Zhong W, Liu Y, Miao H, Li Y, Ji M . Sample Size for Assessing Agreement between Two Methods of Measurement by Bland-Altman Method. Int J Biostat. 2016; 12(2). DOI: 10.1515/ijb-2015-0039. View

Chen K, Stotter C, Klestil T, Nehrer S . Artificial Intelligence in Orthopedic Radiography Analysis: A Narrative Review. Diagnostics (Basel). 2022; 12(9). PMC: 9498096. DOI: 10.3390/diagnostics12092235. View

van de Pol G, Verdonschot N, Van Kampen A . The value of the intra-operative clinical mechanical axis measurement in open-wedge valgus high tibial osteotomies. Knee. 2012; 19(6):933-8. DOI: 10.1016/j.knee.2012.02.003. View

Kijowski R, Liu F, Caliva F, Pedoia V . Deep Learning for Lesion Detection, Progression, and Prediction of Musculoskeletal Disease. J Magn Reson Imaging. 2019; 52(6):1607-1619. PMC: 7251925. DOI: 10.1002/jmri.27001. View

Vaishya R, Vijay V, Birla V, Agarwal A . Inter-observer variability and its correlation to experience in measurement of lower limb mechanical axis on long leg radiographs. J Clin Orthop Trauma. 2016; 7(4):260-264. PMC: 5106474. DOI: 10.1016/j.jcot.2016.05.010. View

Hau M, Menon D, Chan R, Chung K, Chau W, Ho K . Two-dimensional/three-dimensional EOS™ imaging is reliable and comparable to traditional X-ray imaging assessment of knee osteoarthritis aiding surgical management. Knee. 2020; 27(3):970-979. DOI: 10.1016/j.knee.2020.01.015. View

Tipton S, Sutherland J, Schwarzkopf R . The Assessment of Limb Length Discrepancy Before Total Hip Arthroplasty. J Arthroplasty. 2015; 31(4):888-92. DOI: 10.1016/j.arth.2015.10.026. View

Pan Y, Chen Q, Chen T, Wang H, Zhu X, Fang Z . Evaluation of a computer-aided method for measuring the Cobb angle on chest X-rays. Eur Spine J. 2019; 28(12):3035-3043. DOI: 10.1007/s00586-019-06115-w. View

Thienpont E, Schwab P, Cornu O, Bellemans J, Victor J . Bone morphotypes of the varus and valgus knee. Arch Orthop Trauma Surg. 2017; 137(3):393-400. DOI: 10.1007/s00402-017-2626-x. View

10.

Regnard N, Lanseur B, Ventre J, Ducarouge A, Clovis L, Lassalle L . Assessment of performances of a deep learning algorithm for the detection of limbs and pelvic fractures, dislocations, focal bone lesions, and elbow effusions on trauma X-rays. Eur J Radiol. 2022; 154:110447. DOI: 10.1016/j.ejrad.2022.110447. View

11.

Stotter C, Klestil T, Chen K, Hummer A, Salzlechner C, Angele P . Artificial intelligence-based analyses of varus leg alignment and after high tibial osteotomy show high accuracy and reproducibility. Knee Surg Sports Traumatol Arthrosc. 2023; 31(12):5885-5895. PMC: 10719140. DOI: 10.1007/s00167-023-07644-0. View

12.

Schwarz G, Simon S, Mitterer J, Frank B, Aichmair A, Dominkus M . Artificial intelligence enables reliable and standardized measurements of implant alignment in long leg radiographs with total knee arthroplasties. Knee Surg Sports Traumatol Arthrosc. 2022; 30(8):2538-2547. DOI: 10.1007/s00167-022-07037-9. View

13.

Kim S, Park Y, Song M, Lim J, Lee H . Reliability and Validity of the Femorotibial Mechanical Axis Angle in Primary Total Knee Arthroplasty: Navigation versus Weight Bearing or Supine Whole Leg Radiographs. Knee Surg Relat Res. 2018; 30(4):326-333. PMC: 6254869. DOI: 10.5792/ksrr.18.028. View

14.

Schwarz G, Simon S, Mitterer J, Huber S, Frank B, Aichmair A . Can an artificial intelligence powered software reliably assess pelvic radiographs?. Int Orthop. 2023; 47(4):945-953. PMC: 10014709. DOI: 10.1007/s00264-023-05722-z. View

15.

Birkenmaier C, Levrard L, Melcher C, Wegener B, Ricke J, Holzapfel B . Distances and angles in standing long-leg radiographs: comparing conventional radiography, digital radiography, and EOS. Skeletal Radiol. 2024; 53(8):1517-1528. PMC: 11194212. DOI: 10.1007/s00256-024-04592-9. View

16.

Knutson G . Anatomic and functional leg-length inequality: a review and recommendation for clinical decision-making. Part I, anatomic leg-length inequality: prevalence, magnitude, effects and clinical significance. Chiropr Osteopat. 2005; 13:11. PMC: 1232860. DOI: 10.1186/1746-1340-13-11. View

17.

Dallora A, Anderberg P, Kvist O, Mendes E, Diaz Ruiz S, Berglund J . Bone age assessment with various machine learning techniques: A systematic literature review and meta-analysis. PLoS One. 2019; 14(7):e0220242. PMC: 6657881. DOI: 10.1371/journal.pone.0220242. View

18.

Li T, Wang Y, Qu Y, Dong R, Kang M, Zhao J . Feasibility study of hallux valgus measurement with a deep convolutional neural network based on landmark detection. Skeletal Radiol. 2021; 51(6):1235-1247. DOI: 10.1007/s00256-021-03939-w. View

19.

Kim S, Nam J, Kim J, Kim J, Ro D . Enhanced deep learning model enables accurate alignment measurement across diverse institutional imaging protocols. Knee Surg Relat Res. 2024; 36(1):4. PMC: 10785531. DOI: 10.1186/s43019-023-00209-y. View

20.

Applebaum A, Nessim A, Cho W . Overview and Spinal Implications of Leg Length Discrepancy: Narrative Review. Clin Orthop Surg. 2021; 13(2):127-134. PMC: 8173231. DOI: 10.4055/cios20224. View