Dataset Information

Artificial-Intelligence-Aided Radiographic Diagnostic of Knee Osteoarthritis Leads to a Higher Association of Clinical Findings with Diagnostic Ratings.

ABSTRACT:

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

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

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

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.

SUBMITTER: Neubauer M

PROVIDER: S-EPMC9917552 | biostudies-literature | 2023 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Artificial-Intelligence-Aided Radiographic Diagnostic of Knee Osteoarthritis Leads to a Higher Association of Clinical Findings with Diagnostic Ratings.

Neubauer Markus M Moser Lukas L Neugebauer Johannes J Raudner Marcus M Wondrasch Barbara B Führer Magdalena M Emprechtinger Robert R Dammerer Dietmar D Ljuhar Richard R Salzlechner Christoph C Nehrer Stefan S

Journal of clinical medicine 20230117 3

<h4>Background</h4>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.<h4>Methods</h4>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) ...[more]

PMID: 36769394

Similar Datasets

Project description:BackgroundAlthough cross-sectional and cohort data suggest that higher serum blood glucose levels in patients with knee osteoarthritis (KOA) are associated with more severe knee symptoms, little is known about the longitudinal relationship between serum blood glucose and knee symptoms, particularly considering central sensitization (CS) comorbidity, which also worsens knee symptoms.MethodsWe evaluated the longitudinal relationship between serum blood glucose and knee symptoms by dividing the cohort of patients with KOA into those with and without CS. We hypothesized that higher serum blood glucose levels would worsen knee symptoms. A total of 297 participants (mean age: 59.6 years; females: 211; average BMI: 23.7 kg/m2) were enrolled in this study. At baseline, plain radiographs of the bilateral knee joints were evaluated according to the Kellgren-Lawrence grade (KLG). All participants exhibited at least a KLG ≥ 2 in each knee. At baseline, fasting blood glucose (FBG) and Central Sensitization Inventory-9 (CSI-9) were evaluated; ≥ 10 points on the CSI-9 was defined as CS+. Knee injury and Osteoarthritis Outcome Score (KOOS) was evaluated at baseline and at 1-year follow-up; the change in KOOS (ΔKOOS) was calculated by subtracting the KOOS at baseline from that at the 1-year follow-up. Multiple linear regression analysis was conducted with ΔKOOS as the dependent variable and FBG at baseline as the independent variable, adjusted for age, sex, BMI, and CSI-9 at baseline.ResultsOf the 297 subjects, 48 (16.2 %) were defined as CS+. In the CS - group, there was no association between FBG levels at baseline and ΔKOOS. In contrast, FBG at baseline was negatively associated with ΔKOOS pain (B = - 0.448; p = 0.003), ADL (B = - 0.438; p = 0.003), and sports (B = - 0.706; p = 0.007).ConclusionsIn patients with radiographic KOA and CS, higher blood glucose levels were associated with deteriorated knee symptoms during the 1-year follow-up. Healthcare providers should pay attention to controlling blood glucose, particularly in patients with KOA and concurrent CS, to mitigate their knee symptoms.Study designRetrospective cohort study (evidence level: III).

Project description:INTRODUCTION: Osteoarthritis (OA) is considered to be a multifactorial and polygenic disease and diagnosis is mainly clinical and radiological. Correlation between radiographic data and clinical status has been reported. However, very few studies, especially in Caucasian people, describe the association between the Kellgren and Lawrence OA grading scale (KL) and genetic alterations to better understand OA etiopathogenesis and susceptibility. In order to update the knee OA grading, in this study we assessed the associations between KL grade, clinical features such as American Knee Society Score (AKSS), age, and polymorphisms in the principal osteoarthritis susceptibility (OS) genes in Sicilian individuals. METHODS: In 66 Sicilian individuals affected by primary knee OA, the clinical and radiographic evaluation was performed using 2 sub-scores of AKSS (knee score (KS) and function score (FS)) and KL. The patients were also classified according to age. Online Mendelian Inheritance in Man (OMIM) and Database of Single Nucleotide Polymorphisms (dbSNP) Short Genetic Variations databases were used to select gene regions containing the following polymorphisms to analyze: FRZB rs288326 and rs7775, MATN3 rs77245812, ASPN D14 repeats, PTHR2 rs76758470, GDF5 rs143383 and DVWA rs11718863. Patient genotypes were obtained using Sanger DNA sequencing analysis. RESULTS: In our cohort of patients a statistical association between the variables analyzed was reported in all associations tested (KL versus KS, FS and age). We observed that a mild to severe OA radiographic grade is related to severe clinical conditions and loss of articular function and that the severity of symptoms increases with age. Concerning the genotyping analysis, our results revealed a significant statistical association between KL grading and GDF5 rs143383 and DVWA rs11718863 genetic alterations. The latter was also associated with a more severe radiographic grade, displaying its predictive role as OA marker progression. Statistically significant association between clinical, radiographic and genetic signs observed, suggests extending the actual grading of knee OA based mainly on X-ray features. CONCLUSIONS: This work represents a multidisciplinary and translational medicine approach to study OA where clinical, radiological, and OS5 and OS6 SNPs evaluation could contribute to better define grading and progression of OA and to the development of new therapies.

Dataset Information

Artificial-Intelligence-Aided Radiographic Diagnostic of Knee Osteoarthritis Leads to a Higher Association of Clinical Findings with Diagnostic Ratings.

Background

Methods

Results

Conclusion

Publications

Artificial-Intelligence-Aided Radiographic Diagnostic of Knee Osteoarthritis Leads to a Higher Association of Clinical Findings with Diagnostic Ratings.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets