Skip to the content

Validated novel software to measure the conspicuity index of lesions in DICOM images

Szczepura, K and Manning, D 2016, 'Validated novel software to measure the conspicuity index of lesions in DICOM images' , in: Proceedings of SPIE Volume 9787. Medical Imaging 2016: Image Perception, Observer Performance, and Technology Assessment , SPIE.

[img] Microsoft Word - Accepted Version
Restricted to Repository staff only

Download (991kB) | Request a copy
[img] PDF - Accepted Version
Restricted to Repository staff only

Download (917kB) | Request a copy

Abstract

Description of purpose A novel software programme and associated Excel spreadsheet has been developed to provide an objective measure of the expected visual detectability of focal abnormalities within DICOM images. Methodology ROIs are drawn around the abnormality, the software then fits the lesion using a least squares method to recognise the edges of the lesion based on the full width half maximum. 180 line profiles are then plotted around the lesion, giving 360 edge profiles. The co-ordinates show in Figure 1 are captured, as well the standard deviation of the pixel values within the background and lesion (representing anatomical noise and lesion noise respectively). An Excel spreadsheet has been developed to allow variables to be calculated, including SNR and CNR. A conspicuity index has also been developed: Results The software has been validated using the GAMMEX ACR CT accreditation phantom, varying mA, kVp and slice thickness (ST) and the results have been found to give a linear response: Conclusion A novel software programme has been validated to allow calculation of many physical properties of lesions. Additionally, a new measure of conspicuity index has been developed for focal lesions. The analysis could be further developed to incorporate reader decision-analysis data and eye-tracking data allowing correlations between physical and perception measures to be made beyond basic CNR calculations. It could also be used as a tool to distinguish between perceptual and cognitive error. Further refinements could lead to measures of the detectability of more diffuse disease features.

Item Type: Book Section
Editors: Abbey, Craig and Kupinski, Matthew
Schools: Schools > School of Health Sciences > Centre for Health Sciences Research
Publisher: SPIE
ISBN: 9781510600225
Funders: Non funded research
Depositing User: Katy Szczepura
Date Deposited: 04 May 2016 12:35
Last Modified: 04 May 2016 12:35
URI: http://usir.salford.ac.uk/id/eprint/38854

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year