A new artificial intelligence program for the automatic evaluation of scoliosis on frontal spinal radiographs: Accuracy, advantages and limitations

Cover Image

Cite item

Full Text

Abstract

BACKGROUND: Scoliosis is one of the most common spinal deformations that are usually diagnosed on frontal radiographs using Cobb’s method. Automatic measurement methods based on artificial intelligence can overcome many drawbacks of the usual method and can significantly save radiologist’s time.

AIM: To analyze the accuracy, advantages, and disadvantages of a newly developed artificial intelligence program for the automatic diagnosis of scoliosis and measurement of Cobb’s angle on frontal radiographs.

MATERIALS AND METHODS: In total, 114 digital radiographs were used to test the agreement of Cobb’s angle measurements between the new automatic method and the radiologist using the Bland–Altman method on Microsoft Excel. A limited clinical accuracy test was also conducted using 120 radiographs. The accuracy of the system in defining the scoliosis grade was evaluated by sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve.

RESULTS: The agreement of Cobb’s angle measurement between the system and the radiologist’s calculation was found mostly in grade 1 and 2 scoliosis. Only 2.8% of the results showed a clinically significant angle variability of >5°. The diagnostic accuracy metrics of the limited clinical trial in City Mariinsky Hospital (Saint Petersburg, Russia) also proved the reliability of the system, with a sensitivity of 0.97, specificity of 0.88, accuracy (general validity) of 0.93, and area under the receiver operating characteristic curve of 0.93.

CONCLUSION: Overall, the artificial intelligence program can automatically and accurately define the scoliosis grade and measure the angles of spinal curvatures on frontal radiographs.

About the authors

Dima Kh. I. Kassab

Saint Petersburg State University

Author for correspondence.
Email: DimaKK87@gmail.com
ORCID iD: 0000-0001-5085-6614
SPIN-code: 4907-7850

MD

Russian Federation, Saint Petersburg

Irina G. Kamyshanskaya

Saint Petersburg State University

Email: irinaka@mail.ru
ORCID iD: 0000-0002-8351-9216
SPIN-code: 2422-5191

MD, Dr. Sci. (Medicine), Assistant Professor

Russian Federation, Saint Petersburg

Stanislau V. Trukhan

Esper LLC

Email: stas.truhan@gmail.com
ORCID iD: 0000-0003-0688-0988
Russian Federation, Tver

References

  1. Negrini S, Donzelli S, Aulisa AG, et al. 2016 SOSORT guidelines: orthopaedic and rehabilitation treatment of idiopathic scoliosis during growth. Scoliosis. 2018;3(13). doi: 10.1186/s13013-017-0145-8
  2. Birchenko NS. On the asymmetry of right and left leg loading in children with scoliosis. Fundamental’nye issledovaniya. 2005;(4):9–12. (In Russ). EDN: IUMILP
  3. Mironov SP, editor. Orthopedics. Clinical Guidelines. Moscow: GEOTAR-Media; 2018. (In Russ).
  4. Kotel’nikov GP, Lartsev YuV, Ryzhov PV. Traumatology and Orthopedics. 2nd edition, revised. Moscow: GEOTAR-Media; 2021. (In Russ).
  5. Lenke CL. Revised glossary of terms [Internet]. Milwaukee, WI: Scoliosis Research Society; c2023 [cited 2023 Jun 05]. Available from: https://www.srs.org/professionals/online-education-and-resources/glossary/revised-glossary-of-terms
  6. O’Brien MF, Kuklo TR, Blanke KM, et al. Radiographic Measurement Manual. Spinal Deformity Study Group. Miami, Florida: Medtronic Sofamor Danek USA; 2008.
  7. Malfair D, Flemming AK, Dvorak MF, et al. Radiographic evaluation of scoliosis: review. American journal of roentgenology. 2010;194(3):8–22. doi: 10.2214/AJR.07.7145
  8. Imhof H, Halpern B, Herneth AM, et al. Direct diagnosis in radiology. Spinal imaging. 2nd ed. Thieme; 2021.
  9. Kaul V, Enslin S, Gross SA. History of artificial intelligence in medicine. Gastrointest Endosc. 2020;92(4):807–812. doi: 10.1016/j.gie.2020.06.040
  10. Padalko MA, Orlov SV, Naumov AM, Nazarikov SI, Lushnikov AA. Automatic system for determining the angles of scoliotic deformity of the human spine. Vestnik IKBFU. Physics, mathematics, and technology. 2019;(3):55–68. EDN: WMPIXX
  11. Lein GA, Nechaeva NS, Mammadova GМ, Smirnov AA, Statsenko MM. Automation analysis X-ray of the spine to objectify the assessment of the severity of scoliotic deformity in idiopathic scoliosis: a preliminary report. Pediatric Traumatology, Orthopaedics and Reconstructive Surgery. 2020;8(3):317–326. EDN: SJOTEB doi: 10.17816/PTORS34150
  12. Khanal B, Dahal L, Adhikari P, et al. Automatic Cobb angle detection using vertanra detector and vertebra corners regression. Cai Y, Wang L, Audette M, et al. Computational methods and clinical applications for spine imaging. Lecture Notes in Computer Science. Switzerland: Springer Nature, 2020. P:81–87.
  13. Horng M, Kuok S, Fu M, et al. Cobb angle measurement of spine from X-Ray images using convolutional neural network. Computational and mathematical methods in medicine. 2019;2019. doi: 10.1155/2019/6357171
  14. Pan Y, Chen Q, Chen T, et al. Evaluation of a computer-aided method for measuring the Cobb angle on chest X-rays. Eur Spine journal. 2019;28(12):3035–3043. doi: 10.1007/s00586-019-06115-w
  15. Patent RUS database registration № 2022620577/ 17.03.2022. Kassab DKhI, Kamyshanskaya IG, Cheremisin VM, Cheremisin AA. A database of spinal radiographs with different degrees of scoliosis (XrScl). (In Russ).

Supplementary files

Supplementary Files
Action
1. JATS XML
Download
2. Fig. 1. Results of the Bland–Altman method. Agreement between measurements of the two methods in radiographs with grade 1 and 2 scoliosis.

Download (206KB)
Indexing metadata
3. Fig. 2. Results of the Bland–Altman method. Agreement between measurements of the two methods in radiographs with grade 4 scoliosis.

Download (227KB)
Indexing metadata
4. Fig. 3. ROC curve confirms the accuracy of the new AI program in defining the scoliosis grade.

Download (71KB)
Indexing metadata
5. Fig. 4. X-ray image showing grade 2 scoliosis. Analysis by the system (left) and by the radiologist (right). The radiologist did not measure the thoracolumbar curve as the lower EV is not shown in the image.

Download (267KB)
Indexing metadata
6. Fig. 5. Errors in vertebral markings on radiographs with grade 0 (normal). In CXR images, poor definition of the vertebral borders may lead to false measurements.

Download (181KB)
Indexing metadata
7. Fig. 6. Errors in vertebral markings on radiographs with grade 0 (normal). Errors in defining the L5 vertebral body may lead to false positive curve detection (left).

Download (166KB)
Indexing metadata
8. Fig. 7. Errors of the system in evaluating images with grade 0 scoliosis. Cobb’s angles measured by the radiologist (left) and the AI system (right). Measurement variability is not significant (1.4°); however, the scoliosis grade is 0 by the doctor and 1 by the system.

Download (270KB)
Indexing metadata
9. Fig. 8. Distribution of angles measured by the AI system on normal X-ray images; 70% of the 5°–6° range.

Download (270KB)
Indexing metadata
10. Fig. 9. Specifications of vertebral marking in radiographs with grade 3 and 4 scoliosis. Errors in detection and numbering of vertebrae caused by the unusual shape of the vertebrae (yellow arrow).

Download (204KB)
Indexing metadata
11. Fig. 10. Grade 3 scoliosis diagnosed by the radiologist (left) and the AI system (right). Significant variability in measuring the lumbar curve (7.8°) did not affect the overall scoliosis grade.

Download (244KB)
Indexing metadata

Copyright (c) 2024 Eco-Vector

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Согласие на обработку персональных данных

 

Используя сайт https://journals.rcsi.science, я (далее – «Пользователь» или «Субъект персональных данных») даю согласие на обработку персональных данных на этом сайте (текст Согласия) и на обработку персональных данных с помощью сервиса «Яндекс.Метрика» (текст Согласия).