Evaluation and Meta Analysis of CTA Scanning Intelligent System Based on CT Images
Year: 2025
Type: conference-paper
Study designs: Systematic Review, Meta-Analysis
Abstract: Computed Tomography Angiography (CTA) is a frequently utilized medical imaging examination method that can explore the vascular system of the human body. Nonetheless, the reading of the CTA images often requires the doctor to have a significant level of profession knowledge and experience, with possible subjectivity and diagnostic errors in the interpretation of the images. As a result, the development of an intelligent system, based on CTA imaging, capable of automatic analysis, evaluation of the images has become a significant area of research. This study performed a meta-analysis of the literature reporting intelligent systems based on CTA images, and a systematic collection and analysis of the outcomes of these studies in the literature has been performed to assess the performance and reliability of these systems. The study employed systematic literature review and screening, as well as statistical analysis for data. Meta-analysis results revealed a sensitivity range for CNN (Convolutional Neural Network) of 6-9, SVM (Support Vector Machine) of 7-10, and decision tree of 5-8. Decision tree based angiography technology revealed the highest level of sensitivity and evaluation. The outcomes of this study suggested that intelligent CTA angiographic imaging technology presents a potentially significant application in the diagnosis and evaluation of vascular system diseases, and act as a strong supporting tool for physicians. However, the intelligent CTA imaging technology, the development, and application will require additional studies and exploration for the improvement of performance and reliability while monitoring medical safety and privacy.
Source: Procedia Computer Science
Author: Zhilu Wang
Institution: Hebei North University
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4411174891
