Artificial-Intelligence-Aided Radiographic Diagnostic of Knee Osteoarthritis Leads to a Higher Association of Clinical Findings with Diagnostic Ratings.
artificial intelligence
clinical severity scores
knee osteoarthritis
knee radiographs
Journal
Journal of clinical medicine
ISSN: 2077-0383
Titre abrégé: J Clin Med
Pays: Switzerland
ID NLM: 101606588
Informations de publication
Date de publication:
17 Jan 2023
17 Jan 2023
Historique:
received:
03
12
2022
revised:
10
01
2023
accepted:
13
01
2023
entrez:
11
2
2023
pubmed:
12
2
2023
medline:
12
2
2023
Statut:
epublish
Résumé
Radiographic knee osteoarthritis (OA) severity and clinical severity are often dissociated. Artificial intelligence (AI) aid was shown to increase inter-rater reliability in radiographic OA diagnosis. Thus, AI-aided radiographic diagnoses were compared against AI-unaided diagnoses with regard to their correlations with clinical severity. Seventy-one DICOMs (m/f = 27:42, mean age: 27.86 ± 6.5) (X-ray format) were used for AI analysis (KOALA software, IB Lab GmbH). Subjects were recruited from a physiotherapy trial (MLKOA). At baseline, each subject received (i) a knee X-ray and (ii) an assessment of five main scores (Tegner Scale (TAS); Knee Injury and Osteoarthritis Outcome Score (KOOS); International Physical Activity Questionnaire; Star Excursion Balance Test; Six-Minute Walk Test). Clinical assessments were repeated three times (weeks 6, 12 and 24). Three physicians analyzed the presented X-rays both with and without AI via KL grading. Analyses of the (i) inter-rater reliability (IRR) and (ii) Spearman's Correlation Test for the overall KL score for each individual rater with clinical score were performed. We found that AI-aided diagnostic ratings had a higher association with the overall KL score and the KOOS. The amount of improvement due to AI depended on the individual rater. AI-guided systems can improve the ratings of knee radiographs and show a stronger association with clinical severity. These results were shown to be influenced by individual readers. Thus, AI training amongst physicians might need to be increased. KL might be insufficient as a single tool for knee OA diagnosis.
Sections du résumé
BACKGROUND
BACKGROUND
Radiographic knee osteoarthritis (OA) severity and clinical severity are often dissociated. Artificial intelligence (AI) aid was shown to increase inter-rater reliability in radiographic OA diagnosis. Thus, AI-aided radiographic diagnoses were compared against AI-unaided diagnoses with regard to their correlations with clinical severity.
METHODS
METHODS
Seventy-one DICOMs (m/f = 27:42, mean age: 27.86 ± 6.5) (X-ray format) were used for AI analysis (KOALA software, IB Lab GmbH). Subjects were recruited from a physiotherapy trial (MLKOA). At baseline, each subject received (i) a knee X-ray and (ii) an assessment of five main scores (Tegner Scale (TAS); Knee Injury and Osteoarthritis Outcome Score (KOOS); International Physical Activity Questionnaire; Star Excursion Balance Test; Six-Minute Walk Test). Clinical assessments were repeated three times (weeks 6, 12 and 24). Three physicians analyzed the presented X-rays both with and without AI via KL grading. Analyses of the (i) inter-rater reliability (IRR) and (ii) Spearman's Correlation Test for the overall KL score for each individual rater with clinical score were performed.
RESULTS
RESULTS
We found that AI-aided diagnostic ratings had a higher association with the overall KL score and the KOOS. The amount of improvement due to AI depended on the individual rater.
CONCLUSION
CONCLUSIONS
AI-guided systems can improve the ratings of knee radiographs and show a stronger association with clinical severity. These results were shown to be influenced by individual readers. Thus, AI training amongst physicians might need to be increased. KL might be insufficient as a single tool for knee OA diagnosis.
Identifiants
pubmed: 36769394
pii: jcm12030744
doi: 10.3390/jcm12030744
pmc: PMC9917552
pii:
doi:
Types de publication
Journal Article
Langues
eng
Références
Cartilage. 2021 Dec;13(1_suppl):957S-965S
pubmed: 31762295
Best Pract Res Clin Rheumatol. 2016 Aug;30(4):653-669
pubmed: 27931960
Lancet. 2019 Apr 27;393(10182):1745-1759
pubmed: 31034380
Ann Phys Rehabil Med. 2016 Jun;59(3):134-138
pubmed: 26904959
Rheumatol Int. 1992;12(1):33-8
pubmed: 1598499
Radiol Med. 2022 Apr;127(4):398-406
pubmed: 35262842
Ann Rheum Dis. 1957 Dec;16(4):494-502
pubmed: 13498604
Am J Sports Med. 2013 Mar;41(3):590-5
pubmed: 23324431
Arthritis Care Res (Hoboken). 2011 Nov;63 Suppl 11:S208-28
pubmed: 22588746
Osteoarthritis Cartilage. 2014 Jul;22(7):969-74
pubmed: 24857977
Drugs. 1996;52 Suppl 3:1-13
pubmed: 8911794
J Rehabil Med. 2011 Apr;43(5):374-81
pubmed: 21448553
Sensors (Basel). 2022 Mar 10;22(6):
pubmed: 35336346
Eur Radiol. 2019 Apr;29(4):1637-1639
pubmed: 30734848
Osteoarthritis Cartilage. 2013 Sep;21(9):1145-53
pubmed: 23973124
SICOT J. 2022;8:14
pubmed: 35389338
Eur Radiol. 2021 Aug;31(8):6021-6029
pubmed: 33587154
Am J Epidemiol. 1988 Jul;128(1):179-89
pubmed: 3381825
Sci Rep. 2022 Feb 24;12(1):3155
pubmed: 35210490
J Orthop Sports Phys Ther. 2006 Dec;36(12):911-9
pubmed: 17193868
Arthritis Care Res (Hoboken). 2015 Oct;67(10):1379-86
pubmed: 25939659
Arthritis Rheum. 1987 Aug;30(8):914-8
pubmed: 3632732
Knee Surg Sports Traumatol Arthrosc. 2015 Dec;23(12):3532-9
pubmed: 25079135
Osteoarthritis Cartilage. 2008 Jul;16(7):742-8
pubmed: 18417373
Osteoarthritis Cartilage. 1999 Mar;7(2):239-46
pubmed: 10222223
Cartilage. 2017 Jul;8(3):272-282
pubmed: 28618873
Curr Opin Rheumatol. 2023 Jan 1;35(1):44-54
pubmed: 36377320
Bone. 2012 Aug;51(2):278-88
pubmed: 22155587
Arthroplasty. 2022 Mar 5;4(1):16
pubmed: 35246270
J Bone Joint Surg Am. 2014 Jul 16;96(14):1145-1151
pubmed: 25031368
Br Med J (Clin Res Ed). 1982 May 29;284(6329):1607-8
pubmed: 6805625
Ann Rheum Dis. 2008 Jul;67(7):1034-6
pubmed: 18198197
Arthritis Rheum. 2005 Dec;52(12):3918-27
pubmed: 16320339
Arthritis Res Ther. 2012 Jun 07;14(3):212
pubmed: 22682469
PET Clin. 2019 Jan;14(1):17-29
pubmed: 30420217
Osteoarthritis Cartilage. 2007;15 Suppl A:A1-56
pubmed: 17320422
Comput Methods Programs Biomed. 2022 Jul;222:106963
pubmed: 35752117
Clin Orthop Relat Res. 1985 Sep;(198):43-9
pubmed: 4028566
Public Health Nutr. 2006 Sep;9(6):755-62
pubmed: 16925881
Ann Rheum Dis. 2014 Jul;73(7):1323-30
pubmed: 24553908