Multi-Modal Brain Tumor Detection Using Deep Neural Network and Multiclass SVM

Medicina, Год журнала: 2022, Номер 58(8), С. 1090 - 1090

Опубликована: Авг. 12, 2022

Background and Objectives: Clinical diagnosis has become very significant in today's health system. The most serious disease the leading cause of mortality globally is brain cancer which a key research topic field medical imaging. examination prognosis tumors can be improved by an early precise based on magnetic resonance For computer-aided methods to assist radiologists proper detection tumors, imagery must detected, segmented, classified. Manual tumor monotonous error-prone procedure for radiologists; hence, it important implement automated method. As result, classification method presented. Materials Methods: proposed five steps. In first step, linear contrast stretching used determine edges source image. second custom 17-layered deep neural network architecture developed segmentation tumors. third modified MobileNetV2 feature extraction trained using transfer learning. fourth entropy-based controlled was along with multiclass support vector machine (M-SVM) best features selection. final M-SVM classification, identifies meningioma, glioma pituitary images. Results: demonstrated BraTS 2018 Figshare datasets. Experimental study shows that outperforms other both visually quantitatively, obtaining accuracy 97.47% 98.92%, respectively. Finally, we adopt eXplainable Artificial Intelligence (XAI) explain result. Conclusions: Our approach outperformed prior methods. These findings demonstrate obtained higher performance terms enhanced quantitative evaluation accuracy.

Язык: Английский

---

Brain tumor segmentation of MRI images: A comprehensive review on the application of artificial intelligence tools

Computers in Biology and Medicine, Год журнала: 2022, Номер 152, С. 106405 - 106405

Опубликована: Дек. 7, 2022

Язык: Английский

---

Classifying Brain Tumors on Magnetic Resonance Imaging by Using Convolutional Neural Networks

Electronics, Год журнала: 2023, Номер 12(4), С. 955 - 955

Опубликована: Фев. 14, 2023

The study of neuroimaging is a very important tool in the diagnosis central nervous system tumors. This paper presents evaluation seven deep convolutional neural network (CNN) models for task brain tumor classification. A generic CNN model implemented and six pre-trained are studied. For this proposal, dataset utilized Msoud, which includes Fighshare, SARTAJ, Br35H datasets, containing 7023 MRI images. magnetic resonance imaging (MRI) belongs to four classes, three tumors, including Glioma, Meningioma, Pituitary, one class healthy brains. trained with input images several preprocessing strategies applied paper. evaluated Generic CNN, ResNet50, InceptionV3, InceptionResNetV2, Xception, MobileNetV2, EfficientNetB0. In comparison all models, best was obtained an average Accuracy 97.12%. development these techniques could help clinicians specializing early detection

Язык: Английский

---

Brain tumor detection from images and comparison with transfer learning methods and 3-layer CNN

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Фев. 1, 2024

Abstract Health is very important for human life. In particular, the health of brain, which executive vital resource, important. Diagnosis provided by magnetic resonance imaging (MRI) devices, help decision makers in critical organs such as brain health. Images from these devices are a source big data artificial intelligence. This enables high performance image processing classification problems, subfield this study, we aim to classify tumors glioma, meningioma, and pituitary tumor MR images. Convolutional Neural Network (CNN) CNN-based inception-V3, EfficientNetB4, VGG19, transfer learning methods were used classification. F-score, recall, imprinting accuracy evaluate models. The best result was obtained with VGG16 98%, while F-score value same model 97%, Area Under Curve (AUC) 99%, recall precision 98%. CNN architecture models early diagnosis rapid treatment diseases.

Язык: Английский

---

Feature-enhanced deep learning technique with soft attention for MRI-based brain tumor classification

International Journal of Information Technology, Год журнала: 2024, Номер 16(3), С. 1617 - 1626

Опубликована: Янв. 20, 2024

Язык: Английский

---

Advancements and Prospects of Machine Learning in Medical Diagnostics: Unveiling the Future of Diagnostic Precision

Archives of Computational Methods in Engineering, Год журнала: 2024, Номер unknown

Опубликована: Июнь 26, 2024

Язык: Английский

---

Breast tumor localization and segmentation using machine learning techniques: Overview of datasets, findings, and methods

Computers in Biology and Medicine, Год журнала: 2022, Номер 152, С. 106443 - 106443

Опубликована: Дек. 19, 2022

Язык: Английский

---

Application of Deep Neural Networks and Machine Learning algorithms for diagnosis of Brain tumour

2022 International Conference on Computational Intelligence and Sustainable Engineering Solutions (CISES), Год журнала: 2023, Номер unknown, С. 106 - 111

Опубликована: Апрель 28, 2023

Brain tumours are regarded as a fatal condition that impacts the lives of so many people worldwide. The kind, location, and size brain tumour all affect how it will be treated. Hence, an automated diagnosis is needed for early detection. Convolutional neural networks (CNNs) have become increasingly desired in recent times tasks like these. In this work, we performed multi-class classification into different types over MRI scans. For these several convolutional used comparative analysis made better These include AlexNet, GoogleNet, VGG-19, Customized model, ensemble ML models. This certain parameters such optimization, learning Rate, count epochs, Loss. models gave promising results. were evaluated accuracy, F1-Score, Recall, Precision.

Язык: Английский

---

A model for multi-attack classification to improve intrusion detection performance using deep learning approaches

Measurement Sensors, Год журнала: 2023, Номер 30, С. 100924 - 100924

Опубликована: Окт. 21, 2023

This proposed model introduces novel deep learning methodologies. The objective here is to create a reliable intrusion detection mechanism help identify malicious attacks. Deep based solution framework developed consisting of three approaches. first approach Long-Short Term Memory Recurrent Neural Network (LSTM-RNN) with seven optimizer functions such as adamax, SGD, adagrad, adam, RMSprop, nadam and adadelta. evaluated on NSL-KDD dataset classified multi attack classification. has outperformed adamax in terms accuracy, rate low false alarm rate. results LSTM-RNN compared existing shallow machine models methodology (RNN), (LSTM-RNN), (DNN). are bench mark datasets KDD'99, NSL-KDD, UNSWNB15 datasets. self-learnt the features classifies classes multi-attack RNN, provide considerable performance other methods KDD'99 dataset.

Язык: Английский

---

Moving object detection using modified GMM based background subtraction

Measurement Sensors, Год журнала: 2023, Номер 30, С. 100898 - 100898

Опубликована: Окт. 23, 2023

Academics have become increasingly interested in creating cutting-edge technologies to enhance Intelligent Video Surveillance (IVS) performance terms of accuracy, speed, complexity, and deployment. It has been noted that precise object detection is the only way for IVS function well higher level applications including event interpretation, tracking, classification, activity recognition. Through use techniques, current study seeks improve accuracy techniques based on Gaussian Mixture Models (GMM). achieved by developing crucial phases detecting process. In this study, it discussed how model each pixel as a mixture Gaussians update using an online k-means approximation. The adaptive model's distributions are then analyzed identify which ones more likely be product background Each categorized according whether thought include distribution best depicts it.

Язык: Английский

---