Enhancing brain tumor classification with transfer learning: Leveraging DenseNet121 for accurate and efficient detection
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY(2024)
Abstract
Brain tumors pose a serious neurological threat to human life, necessitating improved detection and classification methods. Deep transfer learning (TL), in particular in key tumor categories such as meningioma, pituitary, glioma, and instances without tumors, has shown to be a new and successful method for tumor identification and classification. In this work, the efficacy of two pre-trained TL methods-Inceptionv3 and DenseNet121-was examined for correctly classifying certain kinds of brain tumors. The experimental findings show that the DenseNet-121 model, using the TL approach, performed better than other models in terms of accuracy for the identification and classification of brain tumors. The classification test results were impressive, with DenseNet-121 reaching an astounding 99.95% accuracy and precision, recall, and F1-measure scores of 97.7%, 92.1%, and 94.8%, respectively. DenseNet-121 demonstrated 100% and 92.42% training and validation accuracies, respectively, highlighting its potential as an effective and precise diagnosis tool for brain malignancies.
MoreTranslated text
Key words
brain tumor classification,deep learning,DenseNet-121,Inception V3,transfer learning
AI Read Science
Must-Reading Tree
Example
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined