Brain tumor detection based on extreme learning

Sharif, Muhammad; Amin, Javeria; Raza, Mudassar; Anjum, Muhammad Almas; Afzal, Humaira; Shad, Shafqat Ali

doi:10.1007/s00521-019-04679-8

Cited by 88 publications

(39 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Features were extracted from enhanced images that were used both for training the CNN and performing the classification. Other methods were also introduced in the literature for brain tumor classification, such as a generative adversarial network (GAN)-based approach [19], artificial neural network (ANN)-based learning [27], ELM-based learning [28], residual network [29], standard-features-based classification [30,31], adaptive independent subspace analysis [32], transfer learning-based tumors classification [33], and Excitation DNN [34]. In addition, Togaçar et al [35] proposed a hybrid method based on CNN and feature selection, for the classification of brain tumors.…”

Section: Related Workmentioning

confidence: 99%

Multimodal Brain Tumor Classification Using Deep Learning and Robust Feature Selection: A Machine Learning Application for Radiologists

et al. 2020

View full text Add to dashboard Cite

Manual identification of brain tumors is an error-prone and tedious process for radiologists; therefore, it is crucial to adopt an automated system. The binary classification process, such as malignant or benign is relatively trivial; whereas, the multimodal brain tumors classification (T1, T2, T1CE, and Flair) is a challenging task for radiologists. Here, we present an automated multimodal classification method using deep learning for brain tumor type classification. The proposed method consists of five core steps. In the first step, the linear contrast stretching is employed using edge-based histogram equalization and discrete cosine transform (DCT). In the second step, deep learning feature extraction is performed. By utilizing transfer learning, two pre-trained convolutional neural network (CNN) models, namely VGG16 and VGG19, were used for feature extraction. In the third step, a correntropy-based joint learning approach was implemented along with the extreme learning machine (ELM) for the selection of best features. In the fourth step, the partial least square (PLS)-based robust covariant features were fused in one matrix. The combined matrix was fed to ELM for final classification. The proposed method was validated on the BraTS datasets and an accuracy of 97.8%, 96.9%, 92.5% for BraTs2015, BraTs2017, and BraTs2018, respectively, was achieved.

show abstract

Section: Related Workmentioning

confidence: 99%

Multimodal Brain Tumor Classification Using Deep Learning and Robust Feature Selection: A Machine Learning Application for Radiologists

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Multi-path CNN can extract different features from different modalities. In 2019 and 2020, Muhammad Sharif and Javaria Amin et al proposed several brain tumor segmentation algorithms [18][19][20][21] to further improve the segmentation accuracy and reduce the processing time. Our design is based on the 3D U-Net network proposed in [22], which extends the U-Net network [23] and replaces all 2D operations with 3D operations.…”

Section: Related Workmentioning

confidence: 99%

MRI-based brain tumor segmentation using FPGA-accelerated neural network

Xiong

Fan

et al. 2021

BMC Bioinformatics

View full text Add to dashboard Cite

Background Brain tumor segmentation is a challenging problem in medical image processing and analysis. It is a very time-consuming and error-prone task. In order to reduce the burden on physicians and improve the segmentation accuracy, the computer-aided detection (CAD) systems need to be developed. Due to the powerful feature learning ability of the deep learning technology, many deep learning-based methods have been applied to the brain tumor segmentation CAD systems and achieved satisfactory accuracy. However, deep learning neural networks have high computational complexity, and the brain tumor segmentation process consumes significant time. Therefore, in order to achieve the high segmentation accuracy of brain tumors and obtain the segmentation results efficiently, it is very demanding to speed up the segmentation process of brain tumors. Results Compared with traditional computing platforms, the proposed FPGA accelerator has greatly improved the speed and the power consumption. Based on the BraTS19 and BraTS20 dataset, our FPGA-based brain tumor segmentation accelerator is 5.21 and 44.47 times faster than the TITAN V GPU and the Xeon CPU. In addition, by comparing energy efficiency, our design can achieve 11.22 and 82.33 times energy efficiency than GPU and CPU, respectively. Conclusion We quantize and retrain the neural network for brain tumor segmentation and merge batch normalization layers to reduce the parameter size and computational complexity. The FPGA-based brain tumor segmentation accelerator is designed to map the quantized neural network model. The accelerator can increase the segmentation speed and reduce the power consumption on the basis of ensuring high accuracy which provides a new direction for the automatic segmentation and remote diagnosis of brain tumors.

show abstract

“…It also suggests methods to reduce such symptoms, like Exposure and Response Prevention (ERP). Based on this, researchers can evaluate other inter-related brain diseases as discussed in [26]. A classification system could be designed to leverage this information in order to predict the presence of mental disorders.…”

Section: Table 3 Body Parameter Variation On the Mental State Of A Personmentioning

confidence: 99%

Analysing Mental Health Evaluation Techniques from a Statistical Perspective

Koushik¹,

Deepu²

2021

Atlantis Highlights in Computer Sciences

View full text Add to dashboard Cite

Mental wellbeing is an epitome of success for human beings. Recently, a lot of mental diseases that include anxiety, suicidal tendency, depression, bipolar disorder, obsessive compulsive disorder (OCD), etc. have deep-seated among all age groups. This is due to the fact that the social complexity has increased at work, in relationships, at home and in majorly all aspects of human life. Early detection of these problems can help in treating and if possible, eradicating this illness from a person's life. Researchers have proposed different techniques that can be put into action to effectively identify these diseases. These techniques utilize different activities of normal day-to-day human behaviour, like speech analysis, social media behavioural patterns, visual-activity pattern analysis, etc. Due to a large variety of available techniques, researchers find it difficult to standardize techniques for detection and evaluation of such diseases. Thus, in this paper we have performed an in-depth study about the state-of-the art methods to identify mental health issues. Readers will be able to get a bird's eye view of these methods and will be able to identify the best practices involved in mental health analysis. Moreover, this survey also recommends some possible improvements in the existing methods, in order to further improve the system's performance.

show abstract

Brain tumor detection based on extreme learning

Cited by 88 publications

References 53 publications

Multimodal Brain Tumor Classification Using Deep Learning and Robust Feature Selection: A Machine Learning Application for Radiologists

Multimodal Brain Tumor Classification Using Deep Learning and Robust Feature Selection: A Machine Learning Application for Radiologists

MRI-based brain tumor segmentation using FPGA-accelerated neural network

Analysing Mental Health Evaluation Techniques from a Statistical Perspective

Contact Info

Product

Resources

About