Skip to main content

Advertisement

SpringerLink
Log in

An artificial intelligence based on a convolutional neural network allows a precise analysis of the alignment of the lower limb

  • Original Paper
  • Published:
International Orthopaedics Aims and scope Submit manuscript

Abstract

Purpose

The objective of this study was to develop a numeric tool to automate the analysis of deformity from lower limb telemetry and assess its accuracy. Our hypothesis was that artificial intelligence (AI) algorithm would be able to determine mechanical and anatomical angles to within 1°.

Methods

After institutional review board approval, 1175 anonymized patient telemetries were extracted from a database of more than ten thousand telemetries. From this selection, 31 packs of telemetries were composed and sent to 11 orthopaedic surgeons for analysis. Each surgeon had to identify on the telemetries fourteen landmarks allowing determination of the following four angles: hip-knee-ankle angle (HKA), medial proximal tibial angle (MPTA), lateral distal femoral angle (LDFA), and joint line convergence angle (JLCA). An algorithm based on a machine learning process was trained on our database to automatically determine angles. The reliability of the algorithm was evaluated by calculating the difference of determination precision between the surgeons and the algorithm.

Results

The analysis time for obtaining 28 points and 8 angles per image was 48 ± 12 s for the algorithm. The average difference between the angles measured by the surgeons and the algorithm was around 1.9° for all the angles of interest: 1.3° for HKA, 1.6° for MPTA, 2.1° for LDFA, and 2.4° for JLCA. Intraclass correlation was greater than 95% for all angles.

Conclusion

The algorithm showed high accuracy for automated angle measurement, allowing the estimation of limb frontal alignment to the nearest degree.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

References

  1. Ekeland A, Nerhus TK, Dimmen S et al (2017) Good functional results following high tibial opening-wedge osteotomy of knees with medial osteoarthritis. Knee 24:380–389. https://doi.org/10.1016/j.knee.2016.12.005

    Article  Google Scholar 

  2. Jacquet C, Gulagaci F, Schmidt A et al (2020) Opening wedge high tibial osteotomy allows better outcomes than unicompartmental knee arthroplasty in patients expecting to return to impact sports. Knee Surg Sports Traumatol Arthrosc 28:3849–3857. https://doi.org/10.1007/s00167-020-05857-1

    Article  Google Scholar 

  3. Brinkman J-M, Lobenhoffer P, Agneskirchner JD, et al (2008) Osteotomies around the knee: patient selection, stability of fixation and bone healing in high tibial osteotomies. J Bone Joint Surg Br 90-B:1548–1557 https://doi.org/10.1302/0301-620X.90B12.21198

  4. Paley D, Herzenberg JE, Tetsworth K et al (1994) Deformity planning for frontal and sagittal plane corrective osteotomies. Orthop Clin North Am 25:425–465

    Article  CAS  Google Scholar 

  5. Quirno M, Campbell KA, Singh B et al (2017) Distal femoral varus osteotomy for unloading valgus knee malalignment: a biomechanical analysis. Knee Surg Sports Traumatol Arthrosc 25:863–868. https://doi.org/10.1007/s00167-015-3602-z

    Article  Google Scholar 

  6. Pape D, Hoffmann A, Seil R (2017) Bildgebung und präoperative Planung bei perigenikulären Osteotomien. Oper Orthop Traumatol 29:280–293. https://doi.org/10.1007/s00064-017-0496-6

    Article  CAS  Google Scholar 

  7. Moore J, Mychaltchouk L, Lavoie F (2017) Applicability of a modified angular correction measurement method for open-wedge high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 25:846–852. https://doi.org/10.1007/s00167-015-3954-4

    Article  Google Scholar 

  8. Mirouse G, Dubory A, Roubineau F et al (2017) Failure of high tibial varus osteotomy for lateral tibio-femoral osteoarthritis with < 10° of valgus: outcomes in 19 patients. Orthop Traumatol Surg Res 103:953–958. https://doi.org/10.1016/j.otsr.2017.03.020

    Article  CAS  Google Scholar 

  9. Micicoi G, Jacquet C, Sharma A et al (2021) Neutral alignment resulting from tibial vara and opposite femoral valgus is the main morphologic pattern in healthy middle-aged patients: an exploration of a 3D-CT database. Knee Surg Sports Traumatol Arthrosc Off J ESSKA 29:849–858. https://doi.org/10.1007/s00167-020-06030-4

    Article  Google Scholar 

  10. Schindelin J, Arganda-Carreras I, Frise E et al (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9:676–682. https://doi.org/10.1038/nmeth.2019

    Article  CAS  Google Scholar 

  11. Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional networks for biomedical image segmentation. ArXiv150504597 Cs

  12. Paley D (2002) Normal lower limb alignment and joint orientation. Principles of deformity correction. Springer, Berlin Heidelberg, Berlin, Heidelberg, pp 1–18

    Chapter  Google Scholar 

  13. Hirschmann MT, Hess S, Behrend H et al (2019) Phenotyping of hip-knee-ankle angle in young non-osteoarthritic knees provides better understanding of native alignment variability. Knee Surg Sports Traumatol Arthrosc Off J ESSKA 27:1378–1384. https://doi.org/10.1007/s00167-019-05507-1

    Article  Google Scholar 

  14. Moser LB, Hess S, Amsler F et al (2019) Native non-osteoarthritic knees have a highly variable coronal alignment: a systematic review. Knee Surg Sports Traumatol Arthrosc Off J ESSKA 27:1359–1367. https://doi.org/10.1007/s00167-019-05417-2

    Article  Google Scholar 

  15. Elyasi E, Perrier A, Payan Y (2019) Biomechanical modelling of knee joint for assisting high tibial osteotomy. Proceedings of the 27th Congress of the International Society of Biomechanics (ISB’2019), July 2019, Calgary, Canada

  16. Sohn S, Koh IJ, Kim MS, In Y (2020) Risk factors and preventive strategy for excessive coronal inclination of tibial plateau following medial opening-wedge high tibial osteotomy. Arch Orthop Trauma Surg. https://doi.org/10.1007/s00402-020-03660-8

    Article  Google Scholar 

  17. Miniaci A, Ballmer FT, Ballmer PM, Jakob RP (1989) Proximal tibial osteotomy. A new fixation device. Clin Orthop 250–259

  18. Dugdale TW, Noyes FR, Styer D (1992) Preoperative planning for high tibial osteotomy. The effect of lateral tibiofemoral separation and tibiofemoral length. Clin Orthop 248–264

  19. Feucht MJ, Minzlaff P, Saier T et al (2014) Degree of axis correction in valgus high tibial osteotomy: proposal of an individualised approach. Int Orthop 38:2273–2280. https://doi.org/10.1007/s00264-014-2442-7

    Article  Google Scholar 

  20. Henckel J, Richards R, Lozhkin K, et al (2006) Very low-dose computed tomography for planning and outcome measurement in knee replacement: the imperial knee protocol. J Bone Joint Surg Br 88-B:1513–1518 https://doi.org/10.1302/0301-620X.88B11.17986

  21. Escott BG, Ravi B, Weathermon AC et al (2013) EOS low-dose radiography: a reliable and accurate upright assessment of lower-limb lengths. J Bone Jt Surg 95:e183. https://doi.org/10.2106/JBJS.L.00989

    Article  Google Scholar 

  22. Chernchujit B, Tharakulphan S, Prasetia R et al (2019) Preoperative planning of medial opening wedge high tibial osteotomy using 3D computer-aided design weight-bearing simulated guidance: technique and preliminary result. J Orthop Surg 27:230949901983145. https://doi.org/10.1177/2309499019831455

    Article  Google Scholar 

  23. Guggenberger R, Pfirrmann CWA, Koch PP, Buck FM (2014) Assessment of lower limb length and alignment by biplanar linear radiography: comparison with supine CT and upright full-length radiography. Am J Roentgenol 202:W161–W167. https://doi.org/10.2214/AJR.13.10782

    Article  Google Scholar 

  24. Babazadeh S, Dowsey MM, Bingham RJ et al (2013) The long leg radiograph is a reliable method of assessing alignment when compared to computer-assisted navigation and computer tomography. Knee 20:242–249. https://doi.org/10.1016/j.knee.2012.07.009

    Article  Google Scholar 

  25. Tardy N, Steltzlen C, Bouguennec N et al (2020) Is patient-specific instrumentation more precise than conventional techniques and navigation in achieving planned correction in high tibial osteotomy? Orthop Traumatol Surg Res 106:S231–S236. https://doi.org/10.1016/j.otsr.2020.08.009

    Article  Google Scholar 

  26. Vasta S, Zampogna B, Uribe-Echevarria Marbach B, et al (2019) Correlation of pre-operative planning to surgical correction of opening wedge HTO: a radiographic study utilizing a manual measurement method. J Biol Regul Homeost Agents 33:187–193 XIX Congresso Nazionale S.I.C.O.O.P. Societa’ Italiana Chirurghi Ortopedici Dell’ospedalita’ Privata Accreditata

  27. Blackburn J, Ansari A, Porteous A, Murray J (2018) Reliability of two techniques and training level of the observer in measuring the correction angle when planning a high tibial osteotomy. Knee 25:130–134. https://doi.org/10.1016/j.knee.2017.11.007

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Institute of Movement and Locomotion, Department of Orthopedics and Traumatology, Sainte-Marguerite Hospital, APHM, CNRS, ISM, Aix Marseille University, 270 Boulevard Sainte Marguerite, BP 29, 13274, Marseille, France

    Florent Bernard de Villeneuve, Christophe Jacquet, Jean-Noel Argenson & Matthieu Ollivier

  2. Newclip Technics, Haute-Goulaine, France

    Bilal El Kadim, Mathias Donnez, Olivier Coue & Thibault Poujade

  3. Guy’s & St Thomas’ Hospitals, London, UK

    Raghbir Khakha

  4. Institute for Movement and Locomotion, Hôpital Sainte-Marguerite, Aix-Marseille Université, 270, Boulevard Sainte-Marguerite, 13009, Marseille, France

    Matthieu Ollivier

Authors
  1. Florent Bernard de Villeneuve
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Christophe Jacquet
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Bilal El Kadim
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Mathias Donnez
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Olivier Coue
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Thibault Poujade
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Raghbir Khakha
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Jean-Noel Argenson
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Matthieu Ollivier
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Matthieu Ollivier.

Ethics declarations

Ethical approval

Ethical approval was obtained from review board approval.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Conflict of interest

MO is an educational consultant for Newclip Technics.

JNA is an educational consultant for Zimmer Biomet.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bernard de Villeneuve, F., Jacquet, C., El Kadim, B. et al. An artificial intelligence based on a convolutional neural network allows a precise analysis of the alignment of the lower limb. International Orthopaedics (SICOT) 47, 511–518 (2023). https://doi.org/10.1007/s00264-022-05634-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00264-022-05634-4

Keywords

Search

Navigation