This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Musculoskeletal system, Musculoskeletal soft tissue, Professional issues, Ultrasound, Observer performance, Perception image, Diagnostic procedure, Education and training, Image verification
Authors:
A. F. Abrantes, I. Rodrigues, R. P. P. Almeida, J. P. Pinheiro, O. Lesyuk; Faro/PT
DOI:
10.1594/ecr2016/B-0713
Methods and materials
This is a quantitative and descriptive correlational study.
Data was obtained through a questionnaire in digital format applied directly to a representative sample of radiographers and radiology students.
We relate several dependent variables such as sensitivity,
specificity,
and acuity and response time.
We can also set as independent variables the Study Group (Radiographers and Students),
Supplementary Training and Work Experience.
An Online questionnaire was made with the ViewDEX 2.0 Software in order to make it more convenient and accessible allowing the analysis of 15 MSK images with and without pathology.
The subjects would then respond regarding morphological characteristics and the degree of certainty.
The Software used,
is a standalone application developed in Java Programming Environment.
This program only supports DICOM images,
and lets you configure a new study,
allowing you to create questions and response options for each image.
For the statistical analysis of the data collected,
we used the Statistic Package for Social Sciences (SPSS) and the calculation of some variables through Microsoft Office Excel 2010 software.
First,
the sensitivity (Figure 1) was calculated,
specificity (Figure 2) and acuity (Figure 3),
and was then Cronbach's alpha test to assess the reliability of the questionnaire was applied.
The analysis of qualitative variables typically requires non-parametric methods.
Thus,
we applied the Fisher test.
We used the Kolmogorov-Smirnov to verify the normal distribution of the variable Time and the Levene test for variances of homogeneity.
The Student t test was used to compare the average response time.