Dissertation/Thesis Abstract

Identification of Intercostal Vessels Using Spectral Analysis of Ultrasound Radiofrequency Signals
by Haggard, Asher, M.S., Southern Illinois University at Edwardsville, 2017, 61; 10686414
Abstract (Summary)

In a pilot study, radiofrequency backscatter data was collected in the paravertebral (PV) spaces of 4 healthy individuals. Using the associated gray scale ultrasound and Doppler data as guidance, regions-of-interest (ROIs) were chosen inside vessel in the PV space and outside vessel.. ROI sizes of 1.0 mm, 1.5 mm, and 2.0 mm square were examined for auto-regressive (AR) orders of 10, 20, 30, and 40 and bandwidths of 3dB, 6dB, 20dB. Spectral estimations were performed for each ROI size, AR order, and bandwidth over the A-lines of the ultrasound radiofrequency (RF) data. The spectra were averaged and normalized using data collected from a tissue phantom (Siemens S3000 with a 9L4 probe, Siemens Medical Solutions USA, Inc., Malvern, PA). Eight spectral parameters—Y-intercept, slope, and mid-band fit of the regression line, maximum dB of the spectra, frequency at maximum dB, minimum dB of the spectra, frequency at minimum dB, and integrated backscatter were calculated for each spectral estimate and used create ensembles of bagged tree classifiers. An AR order of 10, bandwidth of 6 dB, and an ROI size of 1.0 mm resulted in the minimum out-of-bag error. An additional random forest, using these chosen values, was created from 70% of the data and evaluated independently with the remaining 30% of data. The random forest achieved a predictive accuracy of 92% and Youden’s Index of 0.85. These results suggest that spectral analysis of ultrasound RF backscatter has the potential to identify intercostal blood vessels.

Using the associated gray scale ultrasound and Doppler data as guidance, regions-of-interest (ROIs) were chosen to represent five tissue types found in and around the PV space—rib, pleura, superior costotransverse ligament, intercostal vessel (artery or vein), and the PV space away from the vessel. An ROI size of 2.0 mm, bandwidth of 20 dB, and AR order 10 had the lowest out-of-bag OOB error at 0.315, and averaged across all tissue types, an accuracy of 89.15%, sensitivity of 0.70, specificity of 0.93, and Youden’s Index (YI) of 0.62. These results show that the identification of the five tissues types in radiofrequency backscatter from intercostal ultrasound is feasible.

Another ROI type was added, rib, along with 2-D spectral parameters, real and complex cepstral peaks were added to the data set. All 5 frames of ultrasound data were used to grow the database by a factor of 5. Ensembles of bagged trees created using ROIs of size 2.0 mm, bandwidths of 20 dB, and AR order of 10 were found to have the lowest OOB error at 0.123. After retraining the classifier for data with an ROI size of 2.0 mm, bandwidth 20, AR order 10, and 200 trees, for the 30% test data, accuracy was found to be an average of 95.90% across all tissue types, average sensitivity was found to be 0.85 across all tissue types, average specificity was found to be 0.98 across all tissue types, and average YI across all tissue types was found to be 0.83.

Indexing (document details)
Advisor: Klingensmith, Jon
Commitee: Noble, Brad, York, Tim
School: Southern Illinois University at Edwardsville
Department: Electrical and Computer Engineering
School Location: United States -- Illinois
Source: MAI 57/02M(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Biomedical engineering, Electrical engineering
Keywords: Intercoastal nerve, Intercoastal vessel, Radiofrequency signals, Spectral analysis, Tissue characterization, Ultrasound
Publication Number: 10686414
ISBN: 9780355592856
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest