Automated segmentation for Breast cancer US image using Projected empirical segmentation
Author: Hasib Md Abid Bin Farid, Dr. Kaisar Alam, Dr. Kazi Khairul Islam
Abstruct: This paper presents a segmentation approach, Projected Empirical Segmentation, designed to identify tissue boundaries in ultrasound video sequences. Starting from an initial manually segmented frame, the method propagates three regions—the interior, exterior, and an intermediate zone requiring classification—through subsequent frames. Statistical features are computed for points within the intermediate zone to determine whether they more closely resemble internal or external tissue characteristics, enabling accurate boundary detection. The process then continues automatically across the video without additional manual input. Experimental evaluation on tendon and blood vessel ultrasound videos demonstrates strong segmentation performance. In this work, we extend and implement the method for breast ultrasound images, highlighting its applicability across different anatomical structures.
Breast Cancer US Image Segmentation and Classification using Mask Region Based Convolutional Neural Network
Author: Hasib Md Abid Bin Farid, Zubair Hasan Chowdhury, Bashir Mahmud Talukder, Sohana Nazneen, Md. Sadman Islam Dipro.
Abstract:Â Breast cancer is one of the most significant global health-related problems among women with the level of occurrence that is steadily increasing. The pathogenesis of this malignancy is supported by a great number of risk factors, including chronological age, family history, genetic predisposition, hormonal processes, the influence of ionizing radiation, obesity, alcohol use, and comprehensive reproductive history. Therefore, timely and accurate diagnosis is required to assure effective treatment procedures and improvement of prognostic outcome. This paper presents a deep learning-based system to segment and classify breast cancer in ultrasound images. In particular, the Mask Region-Based Convolutional Neural Network (Mask R-CNN) is applied to outline and target tumorous areas that can be found in the sonographic images. The main objective of the work is to determine the subtype and anatomical location of breast cancer in the provided ultrasound images. It is expected that the results of this study will support the future research work and also enhance the diagnostic measures in breast cancer detection to improve the potential results.
Segmentation of Skin cancer using Unet Neural Network
Author: Hasib Md Abid Bin Farid, Istiaque Ahmed, Supria Roy, Atika Nawar, Mostofa Habib.
Abstract: The unchecked growth of aberrant cells in the skin leads to skin cancer that can spread to other parts of the body hence causing a great risk to the survival of the patient. These abnormal cells are then identified in time, and thus the most important thing. The recent technological advancements, especially the use of machine learning algorithms on ultrasound imaging have improved the efforts of early detection. However, natural noise and irregularity of the ultrasound images make it difficult to outline the affected regions and this requires the creation of powerful segmentation models and methods. In the current study, we have used U-Net, a dedicated deep-learning network designed to be used in biomedical image segmentation, to extract domains with skin-cancer in ultrasound images. On the basis of platforms like Google Colaboratory and Spyder, we performed segmentation on various test images and compared the results with the respective ground-truth masks in the application of various segmentation methods. We have shown that U-Net achieves good segmentation results. In addition, we have found a positive correlation between accuracy of segmentation and image size as well as training epochs and also batch size, which means that the greater the above parameters the higher the accuracy of segmentation. This paper highlights the effectiveness of U-Net to ultrasound based segmentation of skin-cancer and provides useful information regarding parameter optimization in future studies.
A Fully Automated Segmentation Method Of Ultrasound Images For Cancer Diagnosis Using Modified Empirical Method
Author: Hasib Md Abid Bin Farid, Abdullah Bin Sekander, Shafkut Khan, Md. Mahbubur Rahman, Jaki Abdur Rahman
Abstract: In medical science, ultrasound imaging plays a crucial role in diagnosing tumors, including cancer. Various imaging techniques such as MRI (magnetic resonance imaging) and CT scans (computed tomography) are used to obtain detailed internal views of the human body, and these images are essential for identifying diseases and assessing their severity. However, to extract meaningful diagnostic information, the acquired images must be accurately segmented. Numerous algorithms exist for this segmentation process. The method proposed in this paper aims to improve segmentation accuracy while making the process fully automated by combining multiple techniques including active contour, a modified empirical segmentation approach. Initially, the automated process begins by dividing the image into multiple grids. Active contour is then used for edge-based segmentation, after which the modified empirical segmentation refines the results. Thus, without providing any manual seed, this method can automatically segment a breast cancer image In this work, ultrasound images related to breast cancer are used for evaluation. The performance of the proposed method is assessed by comparing its output with manual segmentations performed by specialists, and the results show a high degree of agreement with expert interpretations.
A Semi Automated Hybrid Segmentation Method For Ultrasound Imaging
Author: Hasib Md Abid Bin Farid, Showrav Sharma Shuvo, Marzia Ashrafi Rasha, Uttam Kumar, Md. Abeer Anjum
Abstract:Â Ultrasound imaging will be critical in contemporary medical diagnosis, especially in the diagnosis of tumors and cancer. Although image guiding techniques like MRI and CT images present useful images of the anatomical regions affected, useful clinical interpretation would be impossible without effective image analysis. There is a great variety of segmentation algorithms with varying degrees of accuracy and efficiency. The paper suggests a semi-automatic system of segmentation that improves the quality of the output by combining the active contours techniques with empirical segmentation. The active contour method in the proposed system will be used as the initializing step in which a preliminary estimate of the optimal boundary is given which is then refined using empirical segmentation. The technique is tested with the help of breast ultrasound images, and its performance is determined by the comparison of the obtained results with the manually generated manual segmentations. The results of the experiment show that the proposed method has a high degree of coincidence with manual analysis, which means its effectiveness and possible use in practice in a clinic.
A Semi Automated Segmentation For Ultrasound Images
Author: Hasib Md Abid Bin Farid, A.S.M. Noor, Zarin Tasnim Nova, Samin Yaser Ayon, S. M. Hasan Shahariyar
Abstract: Medical diagnostics is becoming largely based on a range of imaging modes, including ultrasonography, magnetic resonance imaging (MRI), and computed tomography (CT) to probe areas of concern in the human body. The modalities are essential to detecting a disease and determining the stage of the disease. Minimization of clinically viable information requires the segmentation of such images, a process which has drawn numerous algorithms with different levels of accuracy and computational efficiency. Many of the existing segmentation methodologies are still manual, or have to rely on the supply of several seed points, limiting automation and reproducibility. To overcome these shortcomings, we offer a combination of region-growing and empirical segmentation whereby the former provides an initiation to the latter. This hybrid method is tested on a set of breast ultrasound images, and a measure of its performance is found by comparing the resulting segmentations to those found in a set of expert-performed manual segmentations. According to the experiments, the suggested approach produces credible segmentation results, which proves its possible applicability in the sphere of medical image analysis.