Measurement issues in the sonographic assessment of tennis elbow

Lateral elbow tendinopathy and artificial intelligence: Binary and multilabel findings detection using machine learning algorithms

Background

Ultrasound (US) is a valuable technique to detect degenerative findings and intrasubstance tears in lateral elbow tendinopathy (LET). Machine learning methods allow supporting this radiological diagnosis.

Aim

To assess multilabel classification models using machine learning models to detect degenerative findings and intrasubstance tears in US images with LET diagnosis.

Materials and methods

A retrospective study was performed. US images and medical records from patients with LET diagnosis from January 1st, 2017, to December 30th, 2018, were selected. Datasets were built for training and testing models. For image analysis, features extraction, texture characteristics, intensity distribution, pixel-pixel co-occurrence patterns, and scales granularity were implemented. Six different supervised learning models were implemented for binary and multilabel classification. All models were trained to classify four tendon findings (hypoechogenicity, neovascularity, enthesopathy, and intrasubstance tear). Accuracy indicators and their confidence intervals (CI) were obtained for all models following a K-fold-repeated-cross-validation method. To measure multilabel prediction, multilabel accuracy, sensitivity, specificity, and receiver operating characteristic (ROC) with 95% CI were used.

Results

A total of 30,007 US images (4,324 exams, 2,917 patients) were included in the analysis. The RF model presented the highest mean values in the area under the curve (AUC), sensitivity, and also specificity by each degenerative finding in the binary classification. The AUC and sensitivity showed the best performance in intrasubstance tear with 0.991 [95% CI, 099, 0.99], and 0.775 [95% CI, 0.77, 0.77], respectively. Instead, specificity showed upper values in hypoechogenicity with 0.821 [95% CI, 0.82, −0.82]. In the multilabel classifier, RF also presented the highest performance. The accuracy was 0.772 [95% CI, 0.771, 0.773], a great macro of 0.948 [95% CI, 0.94, 0.94], and a micro of 0.962 [95% CI, 0.96, 0.96] AUC scores were detected. Diagnostic accuracy, sensitivity, and specificity with 95% CI were calculated.

Conclusion

Machine learning algorithms based on US images with LET presented high diagnosis accuracy. Mainly the random forest model shows the best performance in binary and multilabel classifiers, particularly for intrasubstance tears.

Dynamic weight bearing analysis is effective for evaluation of tendinopathy using a customized corridor with multi-directional force sensors in a rat model

Few studies discuss kinetic changes in tendinopathy models. We propose a customized corridor to evaluate dynamic weight bearing (DWB) and shearing forces. Sixty rats were randomly given ultrasound-assisted collagenase injections (Collagenase rats) or needle punctures (Control rats) in their left Achilles tendons, and then evaluated 1, 4, and 8 weeks later. The Collagenase rats always had significantly (p < 0.001) higher histopathological and ultrasound feature scores than did the Controls, significantly lower DWB values in the injured than in the right hindlimbs, and compensatorily higher (p < 0.05) DWB values in the contralateral than in the left forelimbs. The injured hindlimbs had lower outward shearing force 1 and 4 weeks later, and higher (p < 0.05) push-off shearing force 8 weeks later, than did the contralateral hindlimbs. Injured Control rat hindlimbs had lower DWB values than did the contralateral only at week 1. The Collagenase rats had only lower static weight bearing ratios (SWBRs) values than did the Controls at week 1 (p < 0.05). Our customized corridor showed changes in DWB compatible with histopathological and ultrasound feature changes in the rat tendinopathy model. The hindlimb SWBRs did not correspond with any tendinopathic changes.

Ultrasonographic Characteristics of the Common Extensor Tendon of the Elbow in Asymptomatic Individuals: Thickness, Color Doppler Activity, and Bony Spurs

Background:

Ultrasonography (US) of the common extensor tendon (CET) of the elbow is often part of the assessment of patients with lateral epicondylitis. This US assessment is currently based on general tendinopathy references and not well-defined US entities.

Purpose:

To describe CET thickness, color Doppler activity, and bony spurs on US in asymptomatic volunteers and to investigate the influence of sex, age, height, body mass index (BMI), weight, and elbow dominance on the measurements.

Study Design:

Cross-sectional study; Level of evidence, 3.

Methods:

Tendon thickness, color Doppler activity, and bony spurs of the CET were measured sonographically in 264 adults (50% women) aged 20 to 96 years. Two different tendon-thickness measuring techniques were applied, labeled the “plateau measure” and the “1-cm measure.” Color Doppler activity was based on a 0 to 4 rating scale (negative, grades 0 and 1; positive, grades 2-4). A bony spur was defined as a bony outgrowth (≥0.3 mm) arising at the insertional site of the CET.

Results:

With both tendon-thickness measuring techniques, the CET in the dominant elbow was thicker than that in the nondominant elbow, and male tendons were thicker than female tendons (all P ≤ .03). In regression analysis, tendon thickness correlated with weight, color Doppler activity, and arm dominance for both measuring techniques in multiple regression analysis. In addition, the plateau measure correlated with height and the presence of bony spurs. No correlations were observed regarding BMI, sex, or age. Positive color Doppler activity was found in 9% of examined elbows, with no difference between the sexes regarding dominant versus nondominant elbows (all P ≥.20). Bony spurs were found to increase with age, from 23% for people in their 20s to 74% in people older than 70 years. Bony spurs were more common in the dominant elbow (P ≤ .01). Women had a higher prevalence of bony spurs than men, but only in the dominant elbow (P = .03).

Conclusion:

This study presents the US characteristics and normal values of the CET. In 264 asymptomatic participants, the CET was found to be thicker in men and in the dominant elbow. No difference in tendon thickness could be demonstrated with regard to different age groups. Color Doppler activity was found to be positive in nearly 1 of 10 asymptomatic subjects. Bony spurs were a common finding; they increased in prevalence with every decade in age and were considered part of the aging process. Normal variations in CET morphologic characteristics should therefore be considered when implementing US in trials and clinical practice.