Image processing plays a major role in neurologists' clinical diagnosis in the medical field. Several types of imagery are used for diagnostics, tumor segmentation, and classification. Magnetic resonance imaging (MRI) is favored among all modalities due to its noninvasive nature and better representation of internal tumor information. Indeed, early diagnosis may increase the chances of being lifesaving. However, the manual dissection and classification of brain tumors based on MRI is vulnerable to error, time‐consuming, and formidable task. Consequently, this article presents a deep learning approach to classify brain tumors using an MRI data analysis to assist practitioners. The recommended method comprises three main phases: preprocessing, brain tumor segmentation using k‐means clustering, and finally, classify tumors into their respective categories (benign/malignant) using MRI data through a finetuned VGG19 (i.e., 19 layered Visual Geometric Group) model. Moreover, for better classification accuracy, the synthetic data augmentation concept i s introduced to increase available data size for classifier training. The proposed approach was evaluated on BraTS 2015 benchmarks data sets through rigorous experiments. The results endorse the effectiveness of the proposed strategy and it achieved better accuracy compared to the previously reported state of the art techniques.
A brain tumor is an uncontrolled development of brain cells in brain cancer if not detected at an early stage. Early brain tumor diagnosis plays a crucial role in treatment planning and patients' survival rate. There are distinct forms, properties, and therapies of brain tumors. Therefore, manual brain tumor detection is complicated, time-consuming, and vulnerable to error. Hence, automated computer-assisted diagnosis at high precision is currently in demand. This article presents segmentation through Unet architecture with ResNet50 as a backbone on the Figshare data set and achieved a level of 0.9504 of the intersection over union (IoU). The preprocessing and data augmentation concept were introduced to enhance the classification rate. The multi-classification of brain tumors is performed using evolutionary algorithms and reinforcement learning through transfer learning. Other deep learning methods such as ResNet50, DenseNet201, MobileNet V2, and InceptionV3 are also applied. Results thus obtained exhibited that the proposed research framework performed better than reported in state of the art. Different CNN, models applied for tumor classification such as MobileNet V2, Inception V3, ResNet50, Den-seNet201, NASNet and attained accuracy 91.8, 92.8, 92.9, 93.1, 99.6%, respectively. However, NASNet exhibited the highest accuracy.Two processes of transfer learning: freeze and fine-tune, are performed to extract significant features from MRI slices. Brain tumor multi-classification is performed using transfer learning, ResNet50-UNet, and NASNet architecture.
Automatic and precise segmentation and classification of tumor area in medical images is still a challenging task in medical research. Most of the conventional neural network based models usefully connected or convolutional neural networks to perform segmentation and classification. In this research, we present deep learning models using long short term memory (LSTM) and convolutional neural networks (ConvNet) for accurate brain tumor delineation from benchmark medical images. The two different models, that is, ConvNet and LSTM networks are trained using the same data set and combined to form an ensemble to improve the results. We used publicly available MICCAI BRATS 2015 brain cancer data set consisting of MRI images of four modalities T1, T2, T1c, and FLAIR. To enhance the quality of input images, multiple combinations of preprocessing methods such as noise removal, histogram equalization, and edge enhancement are formulated and best performer combination is applied. To cope with the class imbalance problem, class weighting is used in proposed models. The trained models are tested on validation data set taken from the same image set and results obtained from each model are reported. The individual score (accuracy) of ConvNet is found 75% whereas for LSTM based network produced 80% and ensemble fusion produced 82.29% accuracy.
The Internet of Things (IoT) is a recent evolutionary technology that has been the primary focus of researchers for the last two decades. In the IoT, an enormous number of objects are connected together using diverse communications protocols. As a result of this massive object connectivity, a search for the exact service from an object is difficult, and hence the issue of scalability arises. In order to resolve this issue, the idea of integrating the social networking concept into the IoT, generally referred as the Social Internet of Things (SIoT) was introduced. The SIoT is gaining popularity and attracting the attention of the research community due to its flexible and spacious nature. In the SIoT, objects have the ability to find a desired service in a distributed manner by using their neighbors. Although the SIoT technique has been proven to be efficient, heterogeneous devices are growing so exponentially that problems can exist in the search for the right object or service from a huge number of devices. In order to better analyze the performance of services in an SIoT domain, there is a need to impose a certain set of rules on these objects. Our novel contribution in this study is to address the link selection problem in the SIoT by proposing an algorithm that follows the key properties of navigability in small-world networks, such as clustering coefficients, path lengths, and giant components. Our algorithm empowers object navigability in the SIoT by restricting the number of connections for objects, eliminating old links or having fewer connections. We performed an extensive series of experiments by using real network data sets from social networking sites like Brightkite and Facebook. The expected results demonstrate that our algorithm is efficient, especially in terms of reducing path length and increasing the average clustering coefficient. Finally, it reflects overall results in terms of achieving easier network navigation. Our algorithm can easily be applied to a single node or even an entire network.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.