|Title:||Improving diagnostic accuracy for pulmonary embolism using computer-aided detection with artificial neural network approach|
|Subject:||Hong Kong Polytechnic University -- Dissertations|
Pulmonary embolism -- Diagnosis
Diagnostic imaging -- Data processing
Radiography, Medical -- Digital techniques
|Department:||Department of Health Technology and Informatics|
|Pages:||viii, 75 leaves : ill. (some col.) ; 30 cm.|
|Abstract:||Objective: This study aimed to improve diagnosis accuracy of pulmonary embolism (PE) from computer tomography pulmonary angiography by computer-aided detection (CAD) scheme using an artificial neural network (ANN) approach. Methods: A retrospective study was performed for computed tomography pulmonary angiograms requested for clinical suspicious pulmonary embolism during December 2009 to March 2010 in an emergency hospital. Totally 40 image sets, 20 with PE and 20 normal, were selected. A reference standard for presence of PE was established using consensus reading of the same studies by two experienced radiologists. We proposed a CAD scheme implemented with 14 inputs and 2 outputs in predicting the presence or absence of PE. The 14 inputs were decided by the diagnostic parameters for PE through agreement with radiology experts. Then the trained CAD scheme was applied to test using round robin methods for 40 cases (20 normal and 20 abnormal cases). Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of physicians with and without using CAD scheme. Results: The mean area of ROC curve (Az) of physicians without and with CAD was 0.87 and 0.94 respectively. The performance of physicians was improved significantly when the scheme was used (p<0.01). Among other diagnostic parameters, intravascular filling defect was significantly difference between patients with and without PE. Conclusion: Our finding suggested that CAD scheme based on ANN method could improve diagnosis accuracy of PE. We have demonstrated intravascular filling defect and dilatation of right ventricle were reliable parameter for diagnosis of PE. Abbreviations: ROC = Receiver operating characteristic; ANN = artificial neural network; CAD = computer-aided diagnostic; PE = pulmonary embolism; CT = computed tomography; Az= area of the ROC curve|
|Rights:||All rights reserved|
Files in This Item:
|b2411439x.pdf||For All Users (off-campus access for PolyU Staff & Students only)||776.39 kB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: