In the optimization process of lumbar spine examinations, factorial experiments were performed addressing the question of whether the effective dose can be reduced and the image quality maintained by adjusting the image processing parameters. A 2k-factorial design was used which is a systematic and effective method of investigating the influence of many parameters on a result variable. Radiographic images of a Contrast Detail phantom were exposed using the default settings of the process parameters for lumbar spine examinations. The image was processed using different settings of the process parameters. The parameters studied were ROI density, gamma, detail contrast enhancement (DCE), noise compensation, unsharp masking and unsharp masking kernel (UMK). The images were computer analysed and an image quality figure (IQF) was calculated and used as a measurement of the image quality. The parameters with the largest influence on image quality were noise compensation, unsharp masking, unsharp masking kernel and detail contrast enhancement. There was an interaction between unsharp masking and kernel indicating that increasing the unsharp masking improved the image quality when combined with a large kernel size. Combined with a small kernel size however the unsharp masking had a deteriorating effect. Performing a factorial experiment gave an overview of how the image quality was influenced by image processing. By adjusting the level of noise compensation, unsharp masking and kernel, the IQF was improved to a 30% lower effective dose.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1088/0031-9155/52/17/011 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Enhancing mammography: a comprehensive review of computer methods for improving image quality.
Prog Biomed Eng (Bristol)
September 2024
University of Coimbra, CISUC, Department of Informatics Engineering, Coimbra 3030-290, Portugal.
Mammography imaging remains the gold standard for breast cancer detection and diagnosis, but challenges in image quality can lead to misdiagnosis, increased radiation exposure, and higher healthcare costs. This comprehensive review evaluates traditional and machine learning-based techniques for improving mammography image quality, aiming to benefit clinicians and enhance diagnostic accuracy. Our literature search, spanning 2015 - 2024, identified 115 articles focusing on contrast enhancement and noise reduction methods, including histogram equalization, filtering, unsharp masking, fuzzy logic, transform-based techniques, and advanced machine learning approaches.
View Article and Find Full Text PDFDomain Adaptation for Bearing Fault Diagnosis Based on SimAM and Adaptive Weighting Strategy.
Sensors (Basel)
June 2024
School of Electrical Engineering and Automation, Tianjin University of Technology, Tianjin 300384, China.
Domain adaptation techniques are crucial for addressing the discrepancies between training and testing data distributions caused by varying operational conditions in practical bearing fault diagnosis. However, transfer fault diagnosis faces significant challenges under complex conditions with dispersed data and distinct distribution differences. Hence, this paper proposes CWT-SimAM-DAMS, a domain adaptation method for bearing fault diagnosis based on SimAM and an adaptive weighting strategy.
View Article and Find Full Text PDFBrain tumor classification in VIT-B/16 based on relative position encoding and residual MLP.
PLoS One
July 2024
School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan, Hubei, China.
Brain tumors pose a significant threat to health, and their early detection and classification are crucial. Currently, the diagnosis heavily relies on pathologists conducting time-consuming morphological examinations of brain images, leading to subjective outcomes and potential misdiagnoses. In response to these challenges, this study proposes an improved Vision Transformer-based algorithm for human brain tumor classification.
View Article and Find Full Text PDFOptimizing ensemble U-Net architectures for robust coronary vessel segmentation in angiographic images.
Sci Rep
March 2024
Division of Intramural Research, National Institute of Neurological Disorders and Stroke, National Institutes of Health, 35 Convent Dr., Bethesda, MD, 20892, USA.
Article Synopsis
- The study presents an advanced model called SE-RegUNet for accurately segmenting coronary vessels in angiography images, overcoming challenges like uneven contrast and background noise.
- SE-RegUNet utilizes enhanced feature extraction through RegNet encoders and squeeze-and-excitation blocks, along with a unique preprocessing strategy to improve performance, achieving a high Dice score of 0.72 and accuracy of 0.97.
- Although the model shows promise for real-time applications and strong performance on external datasets, further development and clinical trials are needed before it can be widely used in medical settings.
Pulmonary abnormality screening on chest x-rays from different machine specifications: a generalized AI-based image manipulation pipeline.
Eur Radiol Exp
November 2023
Artificial Intelligence Engineering Division, RadiSen Co., Ltd, Seoul, Korea.
Background: Chest x-ray is commonly used for pulmonary abnormality screening. However, since the image characteristics of x-rays highly depend on the machine specifications, an artificial intelligence (AI) model developed for specific equipment usually fails when clinically applied to various machines. To overcome this problem, we propose an image manipulation pipeline.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!