H. N. T. K. Kaldera

Bio: H. N. T. K. Kaldera is an academic researcher from University of Peradeniya. The author has contributed to research in topics: Image segmentation & Convolutional neural network. The author has an hindex of 3, co-authored 4 publications receiving 30 citations.

A Systematic Approach for MRI Brain Tumor Localization and Segmentation Using Deep Learning and Active Contouring.

Abstract: One of the main requirements of tumor extraction is the annotation and segmentation of tumor boundaries correctly. For this purpose, we present a threefold deep learning architecture. First, classifiers are implemented with a deep convolutional neural network (CNN) and second a region-based convolutional neural network (R-CNN) is performed on the classified images to localize the tumor regions of interest. As the third and final stage, the concentrated tumor boundary is contoured for the segmentation process by using the Chan-Vese segmentation algorithm. As the typical edge detection algorithms based on gradients of pixel intensity tend to fail in the medical image segmentation process, an active contour algorithm defined with the level set function is proposed. Specifically, the Chan-Vese algorithm was applied to detect the tumor boundaries for the segmentation process. To evaluate the performance of the overall system, Dice Score, Rand Index (RI), Variation of Information (VOI), Global Consistency Error (GCE), Boundary Displacement Error (BDE), Mean Absolute Error (MAE), and Peak Signal to Noise Ratio (PSNR) were calculated by comparing the segmented boundary area which is the final output of the proposed, against the demarcations of the subject specialists which is the gold standard. Overall performance of the proposed architecture for both glioma and meningioma segmentation is with an average Dice Score of 0.92 (also, with RI of 0.9936, VOI of 0.0301, GCE of 0.004, BDE of 2.099, PSNR of 77.076, and MAE of 52.946), pointing to the high reliability of the proposed architecture.

Brain tumor Classification and Segmentation using Faster R-CNN

Abstract: In this study the problem of fully automated brain tumor classification and segmentation, in Magnetic resonance imaging (MRI) containing both Glioma and Meningioma types of brain tumors are considered. This paper proposes a Convolutional Neural Network (CNN), for classification problem and Faster Region based Convolutional Neural Network (Faster R-CNN) for segmentation problem with reduced number of computations with a higher accuracy level. This research has used 218 images as training set and the systems shows an accuracy of 100% in Meningioma and 87.5% in Glioma classifications and an average confidence level of 94.6% in segmentation of Meningioma tumors. The segmented tumor regions are validated through ground truth analysis and manual analysis by a Neurologist.

MRI based Glioma segmentation using Deep Learning algorithms

Abstract: Primary brain tumors can be malignant (cancerous) or benign (non-cancerous). Out of primary brain tumors, gliomas are the most common and, high grade gliomas carry a poor prognosis. In our paper, we present a technique to segment the glioma cells in Magnetic Resonance Imaging (MRI) using faster Region based Convolutional Neural Network (R-CNN) and edge detection techniques in image processing algorithms. This study identifies the region of interest that is glioma cells, with higher confidence level and localize the tumor on the MRI with the tumor mask. Further, analysis shows that with the proposed technique it is possible to achieve an average detection accuracy, sensitivity, Dice score and confidence level of 99.81%, 87.72%, 91.14% and 93.6% respectively.

A Systematic Approach for MRI Brain Tumor Localization, and Segmentation using Deep Learning and Active Contouring

Abstract: One of the main requirements of tumor extraction is the annotation and segmentation of tumor boundaries correctly. For this purpose, we present a threefold deep learning architecture. First classifiers are implemented with a deep convolutional neural network(CNN) andsecond a region-based convolutional neural network (R-CNN) is performed on the classified images to localize the tumor regions of interest. As the third and final stage, the concentratedtumor boundary is contoured for the segmentation process by using the Chan-Vesesegmentation algorithm. As the typical edge detection algorithms based on gradients of pixel intensity tend to fail in the medical image segmentation process, an active contour algorithm defined with the level set function is proposed. Specifically, Chan- Vese algorithm was applied to detect the tumor boundaries for the segmentation process. To evaluate the performance of the overall system, Dice Score,Rand Index (RI), Variation of Information (VOI), Global Consistency Error (GCE), Boundary Displacement Error (BDE), Mean absolute error (MAE), and Peak Signal to Noise Ratio (PSNR) werecalculated by comparing the segmented boundary area which is the final output of the proposed, against the demarcations of the subject specialists which is the gold standard. Overall performance of the proposed architecture for both glioma and meningioma segmentation is with average dice score of 0.92, (also, with RI of 0.9936, VOI of 0.0301, GCE of 0.004, BDE of 2.099, PSNR of 77.076 and MAE of 52.946), pointing to high reliability of the proposed architecture.

Deep Learning Approach for Brain Tumor Detection and Segmentation

Abstract: Brain tumor is a serious health condition which can be fatal if not treated on time. Hence it becomes necessary to detect the tumor in initial stages for planning treatment at the earliest. In this paper we have proposed a CNN model for detection of brain tumor. Firstly brain MRI images are augmented to generate sufficient data for deep learning. The images are then pre-processed to remove noise and make images suitable for further steps. The proposed system is trained with pre-processed MRI brain images that classifies newly input image as tumorous or normal based on features extracted during training. Back propagation is used while training to minimize the error and generate more accurate results. Autoencoders are used to generated image which removes irrelevant features and further tumor region is segmented using K-Means algorithm which is a unsupervised learning method.

A Novel Deep Learning Method for Recognition and Classification of Brain Tumors from MRI Images.

Abstract: A brain tumor is an abnormal growth in brain cells that causes damage to various blood vessels and nerves in the human body. An earlier and accurate diagnosis of the brain tumor is of foremost important to avoid future complications. Precise segmentation of brain tumors provides a basis for surgical planning and treatment to doctors. Manual detection using MRI images is computationally complex in cases where the survival of the patient is dependent on timely treatment, and the performance relies on domain expertise. Therefore, computerized detection of tumors is still a challenging task due to significant variations in their location and structure, i.e., irregular shapes and ambiguous boundaries. In this study, we propose a custom Mask Region-based Convolution neural network (Mask RCNN) with a densenet-41 backbone architecture that is trained via transfer learning for precise classification and segmentation of brain tumors. Our method is evaluated on two different benchmark datasets using various quantitative measures. Comparative results show that the custom Mask-RCNN can more precisely detect tumor locations using bounding boxes and return segmentation masks to provide exact tumor regions. Our proposed model achieved an accuracy of 96.3% and 98.34% for segmentation and classification respectively, demonstrating enhanced robustness compared to state-of-the-art approaches.

A Comprehensive Review of Indoor/Outdoor Localization Solutions in IoT era: Research Challenges and Future Perspectives

Abstract: The number of connected mobile devices and Internet of Things (IoT) is growing around us, rapidly. Since most of people's daily activities are relying on these connected things or devices. Specifically, this past year (with COVID-19) changed daily life in abroad and this is increased the use of IoT-enabled technologies in the health sector, work, and play. Further, the most common service via using these technologies is the localization/positioning service for different applications including: geo-tagging, billing, contact tracing, health-care system, point-of-interest recommendations, social networking, security, and more. Despite the availability of a large number of localization solutions in the literature, the precision of localization cannot meet the needs of consumers. For that reason, this paper provides an in-depth investigation of the existing technologies and techniques in the localization field, within the IoT era. Furthermore, the benefits and drawbacks of each technique with enabled technologies are illustrated and a comparison between the utilized technologies in the localization is made. The paper as a guideline is also going through all of the metrics that may be used to assess the localization solutions. Finally, the state-of-the-art solutions are examined, with challenges and perspectives regarding indoors/outdoors environments are demonstrated.