Neubauer Markus, Moser Lukas, Neugebauer Johannes, Raudner Marcus, Wondrasch Barbara, Führer Magdalena, Emprechtinger Robert, Dammerer Dietmar, Ljuhar Richard, Salzlechner Christoph, Nehrer Stefan
Danube University Krems, Center for Regenerative Medicine, Dr. Karl-Dorrek-Str. 30, 3500 Krems, Austria.
Karl Landsteiner University of Health Sciences, Department for Orthopedics and Traumatology, University Hospital Krems, Dr. Karl-Dorrek-Straße 30, 3500 Krems, Austria.
J Clin Med. 2023 Jan 17;12(3):744. doi: 10.3390/jcm12030744.
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.
膝关节骨关节炎（OA）的影像学严重程度与临床严重程度常常不相关。研究表明，人工智能（AI）辅助可提高影像学OA诊断中的评分者间信度。因此，就AI辅助的影像学诊断与临床严重程度的相关性，将其与未使用AI的诊断进行了比较。
71份DICOM文件（男/女 = 27:42，平均年龄：27.86 ± 6.5）（X线格式）用于AI分析（KOALA软件，IB Lab GmbH）。研究对象来自一项物理治疗试验（MLKOA）。在基线时，每位受试者接受（i）一次膝关节X线检查和（ii）五项主要评分评估（泰格纳量表（TAS）；膝关节损伤和骨关节炎疗效评分（KOOS）；国际体力活动问卷；星形偏移平衡测试；六分钟步行测试）。临床评估重复三次（第6、12和24周）。三名医生通过KL分级法对提供的X线片分别进行有AI辅助和无AI辅助的分析。对（i）评分者间信度（IRR）和（ii）每位评分者的总体KL评分与临床评分的Spearman相关性检验进行了分析。
我们发现，AI辅助诊断评分与总体KL评分和KOOS的相关性更高。AI带来的改善程度取决于个体评分者。
AI引导系统可提高膝关节X线片的评分，并与临床严重程度表现出更强的相关性。这些结果显示受个体阅片者的影响。因此，可能需要增加医生的AI培训。KL作为膝关节OA诊断的单一工具可能不够充分。
