Carotid Plaque Analysis Software
The main product developed by LifeQ is the Research Software for Ultrasonic Arterial Wall and Athelosclerotic Plaque Measurements. This is a Sophisticated, high powered image analysis package yet designed to be user friendly with the vascular research physician and ultrasonographer in mind.
Main Features included in the package:
- Image Normalization
- Definition of image scale
- Measurements of IMT, plaque size, vessel diameter
- Plaque crop – selection of area of interest
- Extraction of texture features
- Saving of areas of interest as new images
- Saving of measurements and texture features for analysis by Microsoft excel, SPSS or other commercially available statistical packages
- Calculation of the predicted stroke risk based on the ACSRS study results
A user-friendly package is provided, which contains four different modules: (a) image normalization (b) define image scale / measurements (c) standardization of pixel density (d) image crop and (e) extraction of texture features.
Image normalization is performed in order to be able to compare images (Elatrozy et al 1998). Blood and adventitia are used as two reference points and the image is linearly scaled so that zero is the gray scale for blood and 190 for adventitia.
Image Scale Definition Measurements:
This option enables the user to perform several measurements on the ultrasonic images such as Intima-Media Thickness (IMT), plaque thickness, area etc. This option can be used as a stand alone module without using other modules described below. Measurements can be saved on a file for subsequent retrieval and statistical analysis.
This option enables the user to select a plaque (area of interest) and save it as a separate file for use by the features extraction module (see below). Both anterior and posterior wall plaque components can be saved as one image. The program allows for the display of both the gray scale and the color image next to each other. This is particularly useful in the presence of hypoechoic plaques (see front cover).
Plaque Feature Extraction:
The fourth module allows the user to several texture and other image features (fig 1) and store them in a file for subsequent retrieval and analysis by commercially available packages such as Microsoft Excel © and SPSS©. The user can also enter information on the type of plaque, percentage of stenosis, patient status and plaque dark area close to the lumen. The features consists of:
- First order gray level parameters: mean, variance, kurtosis, median, skewness, energy and entropy
- Gray level run length statistics (RUNL): short run emphasis, long run emphasis, gray level distribution, run length distribution, run percentage
- Fourier power spectrum: angular and radial measurements
- Gray-Tone Spatial-Dependence Matrices (SGLDM): angular second moment, contrast, correlation, variance, inverse difference moment, sum average, sum variance, sum entropy, entropy, difference variance, difference entropy information measures of correlation 1 & 2
- Gray level difference statistics (GLDM): homogeneity, contrast, energy, entropy and mean
- Histogram measurements and plaque measurements like area, dark area close to lumen (Kakos et al 2013)
Training for Program Usage:
The correct analysis of images depends on image capturing and normalisation . In order to achieve the best possible results a one day training course on a range of 25 standard images is included in the software price. This will ensure that the operator and his team can obtain reproducible measurements of texture features after image normalization. Additionally the software has a detailed manual with step by step instructions for all modules and instructions to ultrasonographers with recommended settings of duplex scanners.