Medical Ultrasound / Diagnostic Sonography / Ultrasonography
Mahsa Arab; Ali Fallah; Saeid Rashidi; Maryam Mehdizadeh Dastjerdi; Nasrin Ahmadinejad
Volume 17, Issue 2 , September 2023, , Pages 140-150
Abstract
Breast cancer stands as the most prevalent form of cancer among women, with over 80% of early-stage breast abnormalities being benign. Timely detection is therefore crucial for prompt intervention. Ultrasound Radio Frequency (US RF) signals represent a non-invasive, and real-time screening method for ...
Read More
Breast cancer stands as the most prevalent form of cancer among women, with over 80% of early-stage breast abnormalities being benign. Timely detection is therefore crucial for prompt intervention. Ultrasound Radio Frequency (US RF) signals represent a non-invasive, and real-time screening method for breast cancer, offering advantages in tissue differentiation and cost-effectiveness without requiring additional equipment. This research aims to present an intelligent approach for the classification of benign, suspicious, and malignant breast lesions based on effective features extracted from the time series. The dataset, registered as USRFTS, comprises 170 instances recorded from 88 patients. The proposed methodology encompasses four key phases: pre-processing, feature extraction, feature selection, and classification. In the pre-processing phase, B-mode images are reconstructed from US RF time series, and a radiologist manually selects the Region of Interest (ROI) in each image. Subsequently, diverse features in the time and frequency domains are extracted from each ROI during the feature extraction stage. The ant colony method is employed for the selection of impactful features. The dataset is then subjected to evaluation using classifiers such as Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Decision Tree (DT), Linear Discriminant Analysis (LDA), and a reference classification method (RCM). The results demonstrate a maximum classification accuracy of 94.95% for two classes and 93.33% for three classes
Medical Ultrasound / Diagnostic Sonography / Ultrasonography
Saba Jaafari Kia; Hamid Behnam; Majid Vafaeezadeh; Ali Hosseinsabet
Volume 15, Issue 3 , December 2021, , Pages 187-197
Abstract
Heart diseases are main cause factors endangering human health and life, one of the most important heart diseases is valvular heart disease, which has had an increasing trend in recent years. Therefore, if they are diagnosed and treated in time and correctly, they can improve the quality of life and ...
Read More
Heart diseases are main cause factors endangering human health and life, one of the most important heart diseases is valvular heart disease, which has had an increasing trend in recent years. Therefore, if they are diagnosed and treated in time and correctly, they can improve the quality of life and increase the life expectancy, so researchers have always been looking for ways to improve and accelerate the process of diagnosing this disease. Medical images monitoring and recording the activity of the human heart are the main ways to diagnose heart diseases. Processing of these images is generally complex and time consuming, so scientists and experts have always been looking for ways to speed up and facilitate the detection process. Manifold learning is one of the nonlinear dimension reduction methods which has different algorithms and can simplify the processing of echocardiographic images. In this study, using one of the manifold learning algorithms named LLE, we examined echocardiographic images of the heart, and tried to categorize groups with mitral disorders while identifying healthy data from those with disorders. Results show that the method has carefully separated the data of the healthy group from the group with the disorder, and good results were obtained in the data classification. The results show that more than 80% of the samples of the natural group have a different pattern in terms of manifold structure from the samples with the disorder.
Medical Ultrasound / Diagnostic Sonography / Ultrasonography
Hannaneh Keyhanian; Sayed Mahmoud Sakhaei
Volume 12, Issue 3 , November 2018, , Pages 235-248
Abstract
The method of multi-beam beamforming is a low-computational adaptive beamforming method in which, instead of calculating the covariance matrix and inverting it for each point of the image, only one matrix is calculated for all points on the same radial distance. Then, to reduce the complexity of the ...
Read More
The method of multi-beam beamforming is a low-computational adaptive beamforming method in which, instead of calculating the covariance matrix and inverting it for each point of the image, only one matrix is calculated for all points on the same radial distance. Then, to reduce the complexity of the inverse matrix calculation, the problem is solved in the beamspace domain. We introduce a new two-stage method to reduce the complexity of the minimum variance (MV) beamforming method, which outperforms the beamspace method in computational burden aspect in multi-beam method. In the first step, instead of using the signals of all array elements in calculating the covariance matrix, the signals of a decimated one are chosen such that the resulting covariance matrix contains all the correlation information of the signals. In the second stage, the weights of all elements of the array are determined by a proper interpolation method from the weights of the decimated array. According to the simulation results of point targets and cyst phantom, the new method has a performance similar to that of the beamspace multi-beam method in terms of resolution, contrast, and robustness against the errors with at least 3 times lower computational burden.
