Jiří Blahuta

Transcranial Sonography of the Substantia Nigra: Digital Image Analysis

BACKGROUND AND PURPOSE: Increased echogenicity of the substantia nigra is a typical transcranial sonography finding in Parkinson disease. Experimental software for digital analysis of the echogenic substantia nigra area has been developed. The aim of this study was to compare the evaluation of substantia nigra echogenicity by using digital analysis with a manual measurement in patients with Parkinson disease and healthy volunteers. MATERIALS AND METHODS: One hundred thirteen healthy volunteers were enrolled in the derivation cohort, and 50 healthy volunteers and 30 patients with Parkinson disease, in the validation cohort. The substantia nigra was imaged from the right and left temporal bone window by using transcranial sonography. All subjects were examined twice by using different sonographic machines by an experienced sonographer. DICOM images of the substantia nigra were encoded; then, digital analysis and manual measurement of the substantia nigra were performed. The 90th percentile of the derivation cohort values was used as a cut-point for the evaluation of the hyperechogenic substantia nigra in the validation cohort. The Spearman coefficient was used for assessment of the correlation between both measurements. The Cohen κ coefficient was used for the assessment of the correlation between both measurements and Parkinson disease diagnosis. RESULTS: The Spearman coefficient between measurements by using different machines was 0.686 for digital analysis and 0.721 for manual measurement (P < .0001). Hyperechogenic substantia nigra was detected in the same 26 (86.7%) patients with Parkinson disease by using both measurements. Cohen κ coefficients for digital analysis and manual measurement were 0.787 and 0.762, respectively (P < .0001). CONCLUSIONS: The present study showed comparable results when measuring the substantia nigra features conventionally and by using the developed software.

Semi-automatic Ultrasound Medical Image Recognition for Diseases Classification in Neurology

The main aim of this work is semi-automatic ROI positionig in transcranial medical images based on multi-agent systems (MAS) in preprocessing module. Designed approach is based on image processing and is realized by means of artifical intelligence, MAS, which has been experimentally designed in Matlab software environment. Within this processing has been worked with a set of TCS static images in grayscale and binary representation to experimental testing to positioning. This designed application is used for diseases classification in neurology.

An Expert System Based on Using Artificial Neural Network and Region-Based Image Processing to Recognition Substantia Nigra and Atherosclerotic Plaques in B-Images: A Prospective Study

The presented paper is focused on ways of digital image analysis of ultrasound B-images based on echogenicity investigation in determined Region of Interest (ROI). An expert system has been developed in the course of the research. The goal of the paper is to demonstrate how to interconnect automatic finding of the position of the substantia nigra using Artificial Neural Network (ANN) with supervised learning and ROI-based image analysis. For substantia nigra is able to detect the position using ANN from B-image in transverse thalamic plane. From this is computed echogenicity index grade inside the ROI as parkinsonism feature. The methodology is well applicable for a set of images with the same resolution. The results have shown practical application of ANN learning in this case. The second part of the paper is focused on detection of atherosclerotic plaques. An experimental prospective study shown the using ANN can be highly time-consuming problem due to complexity of B-images. The plaques have no standardized shape and size in comparison with SN. To objective appraisal of using ANN to automatic finding atherosclerotic plaque in B-image we need a large set of images of normal and pathological state. Although it is very important using ANN, automatic detection in highly time-consuming problem for ANN training.

Computer analysis and evaluation of ultrasound B-images universally usable due to principle of B-MODE and its displaying in grayscale

We developed an algorithm and simultaneously a computer program to analysis and evaluation of ultrasound B-images. Ultrasound B-imaging is one of mostly used displaying method in radiology. Our developed method and program use of the principle of B-imaging and displaying in grayscale according to pulse-echo approach principle. The program is focused on measurement echogenic area inside defined ROI. So, we measure level of echogenicity for examined tissue. It is based on binary thresholding due to grayscale displaying. We observe how echogenic area is changed according to threshold. Results were verified by an experienced sonographer and also by independent observers. We also observed average correlation between different ultrasound machines (r > 0.82) and between observers (r > 0.85, inter-kappa > 0.83, intra-kappa > 0.9). Up to now we tested the program for substantia nigra echogenicity to Parkinsons Disease diagnosis, raphe nucleus echogenicity and currently we examine atherosclerotic plaques in B-images with high correlation between 2 independent observers (r > 0.75) in a pilot study. Due to reached results of verification we proved the developed algorithm should be universally used in many diagnostics in which is B-MODE used, not only in neurology.