PurposeThe aim of this work is the development and characterization of a model observer (MO) based on convolutional neural networks (CNNs), trained to mimic human observers in image evaluation in terms of detection and localization of low-contrast objects in CT scans acquired on a reference phantom. The final goal is automatic image quality evaluation and CT protocol optimization to fulfill the ALARA principle.ApproachPreliminary work was carried out to collect localization confidence ratings of human observers for signal presence/absence from a dataset of 30,000 CT images acquired on a PolyMethyl MethAcrylate phantom containing inserts filled with iodinated contrast media at different concentrations. The collected data were used to generate the labels for the training of the artificial neural networks. We developed and compared two CNN architectures based respectively on Unet and MobileNetV2, specifically adapted to achieve the double tasks of classification and localization. The CNN evaluation was performed by computing the area under localization-ROC curve (LAUC) and accuracy metrics on the test dataset.ResultsThe mean of absolute percentage error between the LAUC of the human observer and MO was found to be below 5% for the most significative test data subsets. An elevated inter-rater agreement was achieved in terms of S-statistics and other common statistical indices.ConclusionsVery good agreement was measured between the human observer and MO, as well as between the performance of the two algorithms. Therefore, this work is highly supportive of the feasibility of employing CNN-MO combined with a specifically designed phantom for CT protocol optimization programs.
Laser techniques for vibration measurement, due to their non-contact nature, represents an interesting alternative investigational tool to be tested in biomedical and clinic fields. A particular application could be as evaluation method in design and quality control of artificial organs. Aim of this study is to investigate the application of laser vibrometry to the study of mechanical heart valves in-vitro, with an ad hoc set-up. A heterodyne laser Doppler vibrometry system, which allows the measurement of both vibrational velocity and displacement was used. Three different approaches have been carried out, in order to stress the limits of the laser vibrometry technique for testing heart valve prostheses. Critical points and difficulties to build up experimental studies in this field were clearly pointed out. In the present study only one laser head was used, the aim of the authors being to test the feasibility of a simplified approach on mechanical cardiac valves. Starting from that analysis a comparison could be made to assess the capability to discriminate between normal and malfunctioning devices. The advantage of the proposed test bench is that it could provide a non-contact, non-destructive analysis of the valve under the same working conditions as those upon implantation. The proposed method could furnish a typical "fingerprint" characterizing each valve behavior in repeatable experimental conditions.
The measure of the regurgitant flow through heart valves provides an indication of the severity of the valve closure dysfunction with diagnostic relevance. The estimation of the volume passing through the closed valve during systole, and its ratio with the ejection volume, can significantly improve the assessment of an ongoing valvular pathology. The noninvasive quantification of flow converging to the valve is still lacking a satisfying degree of precision. The most popular technique is the Proximal Isovelocity Surface Area (PISA), which assumes that, in the flow field upstream of the valve, the surfaces corresponding to the same velocity are spherical, whence the regurgitant flow is estimated by multiplication with the hemispherical surface area. In the present study, a new method is proposed of color Doppler echocardiography image processing, for regurgitant flow measurements. In this method, called Proximal Arbitrary Surface Conservative Assessment of Leakage (PASCAL), the laws of fluid dynamics are used to reconstruct the entire flow field, in the hypothesis of axial symmetry, starting from the echographic Doppler mapping of one component of velocity. In vitro experiments have confirmed that the new method provides better flow estimates than PISA, on account of its more rigorous physical model of the regurgitant flow.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.