Exploring the Effectiveness of VGG16 and VGG19 Method for Identifying Brain Tumours in MRI Images: A Deep Learning Investigation
Fuente:
Zenodo
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Recurso digital |
| Publicado: |
Zenodo
2024
|
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866901984196427776 |
|---|---|
| author | Darshan P R Prashanth Kumar R Sachidananda M H |
| author_facet | Darshan P R Prashanth Kumar R Sachidananda M H |
| contents | <p><span>An brain tumor illness is brought on by the proliferation of aberrant brain cells. Brain tumors fall into two categories: benign (non-cancerous) and malignant (cancerous) brain tumors. Since brain tumors are uncommon and come in different forms, it is challenging to estimate the chance of surviving of a patient who is tumor-prone. In this work, the context related to medical image analysis, brain tumor identification and categorization have become essential responsibilities. Convolutional Neural Networks (CNNs) have demonstrated impressive implementation in a range of image identification applications thanks to developments in deep learning. Utilising magnetic resonance imaging (MRI) data, we evaluate the performance of two popular CNN architectures, VGG16 and VGG19, in the task of brain tumor detection. Preliminary processing, feature extraction, model training, as well as evaluation are the steps in the multi-step process that makes up the suggested methodology. Pre-processing is done on the MRI scans at first to improve image quality and lower noise. From the pre-processed pictures, high-level features are then extracted using the VGG16 and VGG19 architectures. These designs are made up of multiple layers of convolution that gradually pick up complex characteristics. The findings of VGG 16 so VGG 19 are compared, and the most effective model is employed for forecasting. The precision of the CNN model architecture for VGG16 was found to be 80%, whereas VGG19 had a 98% accuracy rate with a 0.2 loss. Through comparison VGG19 to VGG16 model, we may deduce that it is a trustworthy tool for the quick identification of certain brain tumors.</span></p> <p><strong><span>Keywords</span></strong><span>: Deep Learning, convolution neural networks (CNN) Brain Tumor Detection, Magnetic Resonance Images (MRI), Deep Learning, Artificial Intelligence.</span></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_13639108 |
| institution | Zenodo |
| language | |
| publishDate | 2024 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Exploring the Effectiveness of VGG16 and VGG19 Method for Identifying Brain Tumours in MRI Images: A Deep Learning Investigation Darshan P R Prashanth Kumar R Sachidananda M H <p><span>An brain tumor illness is brought on by the proliferation of aberrant brain cells. Brain tumors fall into two categories: benign (non-cancerous) and malignant (cancerous) brain tumors. Since brain tumors are uncommon and come in different forms, it is challenging to estimate the chance of surviving of a patient who is tumor-prone. In this work, the context related to medical image analysis, brain tumor identification and categorization have become essential responsibilities. Convolutional Neural Networks (CNNs) have demonstrated impressive implementation in a range of image identification applications thanks to developments in deep learning. Utilising magnetic resonance imaging (MRI) data, we evaluate the performance of two popular CNN architectures, VGG16 and VGG19, in the task of brain tumor detection. Preliminary processing, feature extraction, model training, as well as evaluation are the steps in the multi-step process that makes up the suggested methodology. Pre-processing is done on the MRI scans at first to improve image quality and lower noise. From the pre-processed pictures, high-level features are then extracted using the VGG16 and VGG19 architectures. These designs are made up of multiple layers of convolution that gradually pick up complex characteristics. The findings of VGG 16 so VGG 19 are compared, and the most effective model is employed for forecasting. The precision of the CNN model architecture for VGG16 was found to be 80%, whereas VGG19 had a 98% accuracy rate with a 0.2 loss. Through comparison VGG19 to VGG16 model, we may deduce that it is a trustworthy tool for the quick identification of certain brain tumors.</span></p> <p><strong><span>Keywords</span></strong><span>: Deep Learning, convolution neural networks (CNN) Brain Tumor Detection, Magnetic Resonance Images (MRI), Deep Learning, Artificial Intelligence.</span></p> |
| title | Exploring the Effectiveness of VGG16 and VGG19 Method for Identifying Brain Tumours in MRI Images: A Deep Learning Investigation |
| url | https://doi.org/10.5281/zenodo.13639108 |