Brain Tumor Diagnosis Using Machine Learning, Convolutional Neural Networks, Capsule Neural Networks and Vision Transformers, Applied to MRI: A Survey

Akinyelu, Andronicus Ayobami; Zaccagna, Fulvio; Grist, James T.; Castelli, Mauro; Rundo, Leonardo

doi:10.3390/jimaging8080205

Cited by 59 publications

(31 citation statements)

References 108 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2022, Akinyelu et al published a survey in which they compare the most recently developed segmentation techniques based on ML, CNN, Capsule Networks (CapsNet), and Vision Transformers (ViT). Most of these methods are used for segmentation or classification tasks, which are strictly related since they both contribute to identifying the grade of a brain tumor and planning its best treatment [ 92 ].…”

Section: Segmentationmentioning

confidence: 99%

“…Most CNN networks can extract information only from 2D MRI images. However, some recent studies aimed to extract volumetric information in 3D MRI images using CNN models [ 77 , 92 ]. ViT-based models, for example, can be used for 2D and 3D image segmentation and classification.…”

Section: Segmentationmentioning

confidence: 99%

“…CapsNet-based tumor segmentation techniques have been proposed in the literature to address the downsides of CNNs methods. As previously mentioned CapsNets require smaller datasets to be trained in comparison with CNNs and consider the tumor surrounding tissues [ 92 , 93 ]. Even if most of the CapsNet-based techniques proposed in the literature are used for brain tumor classification, CapsNet-based models are also very useful for segmentation tasks, since they need small-scale datasets to train and require lesser computational complexity compared to CNN-based techniques.…”

Section: Segmentationmentioning

confidence: 99%

See 2 more Smart Citations

Artificial Intelligence in Brain Tumor Imaging: A Step toward Personalized Medicine

et al. 2023

View full text Add to dashboard Cite

The application of artificial intelligence (AI) is accelerating the paradigm shift towards patient-tailored brain tumor management, achieving optimal onco-functional balance for each individual. AI-based models can positively impact different stages of the diagnostic and therapeutic process. Although the histological investigation will remain difficult to replace, in the near future the radiomic approach will allow a complementary, repeatable and non-invasive characterization of the lesion, assisting oncologists and neurosurgeons in selecting the best therapeutic option and the correct molecular target in chemotherapy. AI-driven tools are already playing an important role in surgical planning, delimiting the extent of the lesion (segmentation) and its relationships with the brain structures, thus allowing precision brain surgery as radical as reasonably acceptable to preserve the quality of life. Finally, AI-assisted models allow the prediction of complications, recurrences and therapeutic response, suggesting the most appropriate follow-up. Looking to the future, AI-powered models promise to integrate biochemical and clinical data to stratify risk and direct patients to personalized screening protocols.

show abstract

Section: Segmentationmentioning

confidence: 99%

Section: Segmentationmentioning

confidence: 99%

Section: Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Artificial Intelligence in Brain Tumor Imaging: A Step toward Personalized Medicine

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Machine and deep learning algorithms are applied to the MRI images for three major applications, namely, tumor detection, segmentation, and grade estimation. The survey performed in [ 109 ] evidently described four types of brain tumor segmentation & classification techniques, namely, classical machine learning techniques, CNN-based techniques, capsule neural network-based techniques, and vision transformer-based techniques. Different feature extraction methods, namely first-order statistical feature extraction, gray-level co-occurrence matrix, histogram-oriented gradient, etc., have been used to extract the texture information from MRI images, and the texture information is used by multiple machine learning algorithms to classify the tumor [ 110 ].…”

Section: Applications Of Ai For Neurological Disordersmentioning

confidence: 99%

Convergence of Artificial Intelligence and Neuroscience towards the Diagnosis of Neurological Disorders—A Scoping Review

Surianarayanan

Lawrence

Raj

et al. 2023

Sensors

View full text Add to dashboard Cite

Artificial intelligence (AI) is a field of computer science that deals with the simulation of human intelligence using machines so that such machines gain problem-solving and decision-making capabilities similar to that of the human brain. Neuroscience is the scientific study of the struczture and cognitive functions of the brain. Neuroscience and AI are mutually interrelated. These two fields help each other in their advancements. The theory of neuroscience has brought many distinct improvisations into the AI field. The biological neural network has led to the realization of complex deep neural network architectures that are used to develop versatile applications, such as text processing, speech recognition, object detection, etc. Additionally, neuroscience helps to validate the existing AI-based models. Reinforcement learning in humans and animals has inspired computer scientists to develop algorithms for reinforcement learning in artificial systems, which enables those systems to learn complex strategies without explicit instruction. Such learning helps in building complex applications, like robot-based surgery, autonomous vehicles, gaming applications, etc. In turn, with its ability to intelligently analyze complex data and extract hidden patterns, AI fits as a perfect choice for analyzing neuroscience data that are very complex. Large-scale AI-based simulations help neuroscientists test their hypotheses. Through an interface with the brain, an AI-based system can extract the brain signals and commands that are generated according to the signals. These commands are fed into devices, such as a robotic arm, which helps in the movement of paralyzed muscles or other human parts. AI has several use cases in analyzing neuroimaging data and reducing the workload of radiologists. The study of neuroscience helps in the early detection and diagnosis of neurological disorders. In the same way, AI can effectively be applied to the prediction and detection of neurological disorders. Thus, in this paper, a scoping review has been carried out on the mutual relationship between AI and neuroscience, emphasizing the convergence between AI and neuroscience in order to detect and predict various neurological disorders.

show abstract

“…This study covered the methodologies of DL-based diagnostic approaches for PD detection, including PD dataset pre-processing, feature extraction and selection, and classification. Andronicus et al [ 20 ] used capsule neural networks (Caps-Nets) for a revolutionary sort of machine learning (ML) architecture that was recently designed to solve the shortcomings of conventional neural networks (CNNs). Caps-Nets are resistant to rotations and affine translations, which is advantageous for analyzing medical image datasets.…”

Section: Related Workmentioning

confidence: 99%

Grade Classification of Tumors from Brain Magnetic Resonance Images Using a Deep Learning Technique

et al. 2023

View full text Add to dashboard Cite

To improve the accuracy of tumor identification, it is necessary to develop a reliable automated diagnostic method. In order to precisely categorize brain tumors, researchers developed a variety of segmentation algorithms. Segmentation of brain images is generally recognized as one of the most challenging tasks in medical image processing. In this article, a novel automated detection and classification method was proposed. The proposed approach consisted of many phases, including pre-processing MRI images, segmenting images, extracting features, and classifying images. During the pre-processing portion of an MRI scan, an adaptive filter was utilized to eliminate background noise. For feature extraction, the local-binary grey level co-occurrence matrix (LBGLCM) was used, and for image segmentation, enhanced fuzzy c-means clustering (EFCMC) was used. After extracting the scan features, we used a deep learning model to classify MRI images into two groups: glioma and normal. The classifications were created using a convolutional recurrent neural network (CRNN). The proposed technique improved brain image classification from a defined input dataset. MRI scans from the REMBRANDT dataset, which consisted of 620 testing and 2480 training sets, were used for the research. The data demonstrate that the newly proposed method outperformed its predecessors. The proposed CRNN strategy was compared against BP, U-Net, and ResNet, which are three of the most prevalent classification approaches currently being used. For brain tumor classification, the proposed system outcomes were 98.17% accuracy, 91.34% specificity, and 98.79% sensitivity.

show abstract

Brain Tumor Diagnosis Using Machine Learning, Convolutional Neural Networks, Capsule Neural Networks and Vision Transformers, Applied to MRI: A Survey

Cited by 59 publications

References 108 publications

Artificial Intelligence in Brain Tumor Imaging: A Step toward Personalized Medicine

Artificial Intelligence in Brain Tumor Imaging: A Step toward Personalized Medicine

Convergence of Artificial Intelligence and Neuroscience towards the Diagnosis of Neurological Disorders—A Scoping Review

Grade Classification of Tumors from Brain Magnetic Resonance Images Using a Deep Learning Technique

Contact Info

Product

Resources

About