Compensatory Nursing for Patients with Post-Hepatitis B Virus Cirrhosis Through Ultrasound Imaging Information
Objective: To identify and classify the B-mode ultrasound images by utilizing the elastic algorithm of wavelet packet decomposition (WPD) and Bayesian belief network (BBN), thereby increasing the cure rate of cirrhosis by interfering the compensatory nursing for cirrhosis patients
who were previously infected with hepatitis B virus (HBV). Method: First, the WPD-based elastic feature extraction algorithm (FEA), the texture FEA, the time series-based energy spectrum FEA, and the BBN are introduced. Also, the statistical feature distribution of the five fitting
parameters is analyzed. Based on these three FEAs, the classification and recognition experiments of normal liver samples and liver cirrhosis samples are compared to complete the grayscale imaging of B ultrasound images. To further improve the accuracy of imaging, these elastic new features
are used to build a BBN model, and the output posterior probability is used as a new parameter to realize liver ultrasound B-mode imaging. Results: Among all the classification models, the recognition rates based on the elastic new features are the highest. Especially, the average recognition
rate of elastic features under random forest has reached 95%. The elastic parameters extracted from radio frequency (RF) signals obtained by conventional ultrasound diagnostic instruments are used for elastography, which can characterize the elastic information of liver tissue. Based on the
elastic features, a BBN model is constructed. The pseudo-color coding of the output posterior probability of the model can characterize the stiffness of liver tissue. Also, its new parameter elastography intuitively displays the elastic information of liver tissue. Conclusion: The proposed
elastic FEA of WPD and BBN have an excellent ability to identify and classify ultrasound images of cirrhosis, which can be applied in the clinical diagnosis of cirrhosis, as well as the compensatory nursing for patients with post-HBV cirrhosis.
Keywords: Bayesian Belief Network; Cirrhosis; Compensatory Nursing; Elastography; Ultrasound Image
Document Type: Research Article
Affiliations: 1: Department of Infections, First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, 710061, China 2: Department of Ultrasound, First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, 710061, China 3: Malaysian Society for Public Health & Medicine, Putrajaya 62675, Malaysia
Publication date: 01 August 2020
- Journal of Medical Imaging and Health Informatics (JMIHI) is a medium to disseminate novel experimental and theoretical research results in the field of biomedicine, biology, clinical, rehabilitation engineering, medical image processing, bio-computing, D2H2, and other health related areas.
- Editorial Board
- Information for Authors
- Subscribe to this Title
- Ingenta Connect is not responsible for the content or availability of external websites
- Access Key
- Free content
- Partial Free content
- New content
- Open access content
- Partial Open access content
- Subscribed content
- Partial Subscribed content
- Free trial content