Current Issue : October-December Volume : 2022 Issue Number : 4 Articles : 5 Articles
Magnetic resonance imaging (MRI) is an accurate and noninvasive method employed for the diagnosis of various kinds of diseases in medical imaging. Most of the existing systems showed significant performances on small MRI datasets, while their performances decrease against large MRI datasets. Hence, the goal was to design an efficient and robust classification system that sustains a high recognition rate against large MRI dataset. Accordingly, in this study, we have proposed the usage of a novel feature extraction technique that has the ability to extract and select the prominent feature from MRI image. The proposed algorithm selects the best features from the MRI images of various diseases. Further, this approach discriminates various classes based on recursive values such as partial Z-value. The proposed approach only extracts a minor feature set through, respectively, forward and backward recursion models. The most interrelated features are nominated in the forward regression model that depends on the values of partial Z-test, while the minimum interrelated features are diminished from the corresponding feature space under the presence of the backward model. In both cases, the values of Z-test are estimated through the defined labels of the diseases. The proposed model is efficiently looking the localized features, which is one of the benefits of this method. After extracting and selecting the best features, the model is trained by utilizing support vector machine (SVM) to provide the predicted labels to the corresponding MRI images. To show the significance of the proposed model, we utilized a publicly available standard dataset such as Harvard Medical School and Open Access Series of Imaging Studies (OASIS), which contains 24 various brain diseases including normal. The proposed approach achieved the best classification accuracy against existing state-of-the-art systems....
Introduction: Diagnostic Reference Levels (DRLs), typically set at the 75th percentile of the dose distribution from surveys conducted across a broad user base using a specified dose-measurement protocol, are recommended for radiological examinations. There is a need to develop and implement DRLs as a standardisation and optimisation tool for the radiological protection of patients at Computed Tomography (CT) facilities. Methods: This was a retrospective cross-sectional study conducted in seven (7) different CT scan facilities in which participants were recruited by systematic random sampling. The study variables were dose length product (DLP) and volume-weighted CTDI (CTDIvol) for the radiation doses for head, chest, abdomen and lumbar spine CT examinations. The DRLs for CTDIvol and DLP were obtained by calculating the 3rd quartiles of the radiation doses per study site by anatomical region. The national diagnostic reference levels were determined by computation of DRLs using the 75th centile of the median values. Results: A total of 574 patients were examined with an average age of 47.1 years. For CTDIvol estimates; there was a strong positive significant relationship between the CTDIvol and examination mAs (rs = 0.9017, p-value < 0.001), and reference mAs (rs = 0.0.7708, p-value < 0.001). For DLP estimates; there was a moderate positive significant relationships between DLP and total mAs (rs = 0.6812, p-value < 0.001), reference mAs (rs = 0.5493, p-value < 0.001). The DRLs were as follows; for head CT scan – the average median CTDIvol was 56.02 mGy and the DLP was 1260.3 mGy.cm; for Chest CT, the CTDI volume was 7.82 mGy and the DLP was 377.0 mGy.cm; for the abdomen CT, the CTDI volume 12.54 mGy and DLP 1418.3 mGy.cm and for the lumbar spine 19.48 mGy and the DLP was 843 mGy.cm, respectively. Conclusion: This study confirmed the need to optimize the CT scan parameters in order to lower the national DRLs. This can be achieved by extensive training of all the CT scan radiographers on optimizing the CT scan acquisition parameters. Continuous dose audits are also advised with new equipment or after every three years to ensure that values out of range are either justified or further investigated....
Abstract: Background and Objectives: Cerebral complications related to the COVID-19 were documented by brain MRIs during the acute phase. The purpose of the present study was to describe the evolution of these neuroimaging findings (MRI and FDG-PET/CT) and describe the neurocognitive outcomes of these patients. Methods: During the first wave of the COVID-19 outbreak between March 1 and May 31, 2020, 112 consecutive COVID-19 patients with neurologic manifestations underwent a brain MRI at Strasbourg University hospitals. After recovery, during follow-up, of these 112 patients, 31 (initially hospitalized in intensive care units) underwent additional imaging studies (at least one brain MRI). Results: Twenty-three men (74%) and eight women (26%) with a mean age of 61 years (range: 18–79) were included. Leptomeningeal enhancement, diffuse brain microhemorrhages, acute ischemic strokes, suspicion of cerebral vasculitis, and acute inflammatory demyelinating lesions were described on the initial brain MRIs. During follow-up, the evolution of the leptomeningeal enhancement was discordant, and the cerebral microhemorrhages were stable. We observed normalization of the vessel walls in all patients suspected of cerebral vasculitis. Four patients (13%) demonstrated new complications during follow-up (ischemic strokes, hypoglossal neuritis, marked increase in the white matter FLAIR hyperintensities with presumed vascular origin, and one suspected case of cerebral vasculitis). Concerning the grey matter volumetry, we observed a loss of volume of 3.2% during an average period of approximately five months. During follow-up, the more frequent FDG-PET/CT findings were hypometabolism in temporal and insular regions. Conclusion: A minority of initially severe COVID-19 patients demonstrated new complications on their brain MRIs during follow-up after recovery....
Researchers are continuously exploring the potential use of microwave imaging in the early detection of breast cancer. The technique offers a promising alternative to mammography, a standard clinical imaging procedure today. The contrast in dielectric properties between normal and cancerous tissues makes microwave imaging a viable technique for detecting breast cancer. Experimental results are presented in this paper that demonstrate the detection of breast cancer using microwaves operating at 2.4 GHz. The procedure involves antenna fabrication, phantom tissue development, and image reconstruction. Design and fabrication of patch antenna are used in the study, described in detail. The patch antenna pair is used for transmitting and receiving source waves. Tissue mimicking models were developed from paraffin wax and glycerin for the dielectric constants of 9 and 47, respectively, representing the tissue and tumor. Further, AI-based tomographic images were obtained by implementing a filtered back-projection algorithm in the computer. In the results, the presence of the tumor is quantitatively analyzed....
Breast cancer screening and detection using high-resolution mammographic images have always been a difficult task in computer vision due to the presence of very small yet clinically significant abnormal growths in breast masses. The size difference between such masses and the overall mammogram image as well as difficulty in distinguishing intra-class features of the Breast Imaging Reporting and Database System (BI-RADS) categories creates challenges for accurate diagnosis. To obtain near-optimal results, object detection models should be improved by directly focusing on breast cancer detection. In this work, we propose a new two-stage deep learning method. In the first stage, the breast area is extracted from the mammogram and small square patches are generated to narrow down the region of interest (RoI). In the second stage, breast masses are detected and classified into BI-RADS categories. To improve the classification accuracy for intra-classes, we design an effective tumor classification model and combine its results with the detection model’s classification scores. Experiments conducted on the newly collected high-resolution mammography dataset demonstrate our two-stage method outperforms the original Faster R-CNN model by improving mean average precision (mAP) from 0.85 to 0.94. In addition, comparisons with existing works on a popular INbreast dataset validate the performance of our two-stage model....
Loading....