Article Text

Download PDFPDF
  1. A de-Groot-Ferrando,
  2. J Ríos-Díaz,
  3. J J Martínez-Payá,
  4. M E del-Baño-Aledo
  1. ‘Preventive Ultrasonography, Morpho-densitometry & Research in Physiotherapy and Manual Therapy’ (ECOFISTEM) Research Group, San Antonio Catholic University of Murcia, Spain


    Introduction In recent years, musculoskeletal ultrasound (US) is being introduced as a useful biomedical imaging technique.1 However, there are few works about the analysis and the processing of US image to extract quantitative information.2 ,3 Textural analysis with grey level co-occurrence matrix (gLCM) is based on the comparison of the grey levels of pairs of pixels, which builds a matrix of co-occurrence probability along the entire image.4 The aim of this study was to determine morpho-textural parameters of patellar ligament (PL) in professional volleyball players along one season.

    Table 1.

    Descriptive statistics for statistically significant parameters

    Methods An observational, longitudinal and analytical study was designed with 66 cross-sectional ultrasonograms of PL (n=33 male players; mean age: 27.3 year (S.D.:4.72 year)). We obtained cross-sectional PL ultrasonograms, 5 mm from the distal border of the patella (LogiqE Ultrasound System, Enraf Nonius-General Electric). The records were taken bilaterally at the beginning and at the end of the season (6 months). The morphometric parameters were: echogenicity, echogenicity variation, width, thickness and circularity; and GLCM textural features: contrast and entropy. Image analysis was performed with ImageJ software v1.44. SPSS Statistics Software (V.19.0) was used for analysis of variance (ANOVA) for repeated measures. The size of effect was calculated with Cohen's d. All CIs were fixed at 95%.

    Results The echogenicity (F=0.075; p=0.785; d=0.02) and echogenicity variation (F=3.308; p=0.074; d=0.17) showed no significant changes along the season. However the width was higher (F=6.144; p=0.016) at the end of season with a moderate effect size (d=0.36). The thickness (F=5.67; p=0.02) was lower with a small effect size (d=0.19) and the circularity (F=4.198; p=0.045) also was lower (d=0.17). As for the textural parameters, the contrast (F=19.438; p<0.001) increased at the end of the season with a moderate-high size effect (d=0.54); however, the entropy (F=0.081; p=0.78) did not change. See table 1 at the bottom of the previous page for descriptive statistics for statistically significant parameters.

    Discussion Mechanical stress can generate adaptive changes in tendons. The study of these changes is complicated and invasive in living subjects, but with the advanced image analysis of ultrasonograms, it could be possible in an indirect way to detect these changes. Textural complex algorithms (as GLCM) provide information about the ultrasonography patterns. In this study the contrast that measures the relationship of grey level of pair of pixels was higher at the end of season; this could be related to a higher synthesis of collagen. In future studies, it would be interesting to find the relationship between biochemical markers and textural parameters.

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.