Oct 2026· Journal of Computer & Electrical and Electronics Engineering Sciences· Vol 4, pp. 77-82· 0 citations
TL;DR
MobileNetV2 (pre-trained) architecture has outstanding performance in terms of MRI based neoplasm classification and the combination of the attention mechanisms and Grad-CAM interpretability techniques can improve the accuracy and transparency of the diagnosis.
Abstract
Aims: This paper explores clinical explainability of neural network architectures to classify brain neoplasma using the magnetic resonance imaging patterns.
Methods: We tested various architectures, such as a standard Convolutional Neural Network as our reference network and our attention-augmented CNN model, as well as pre-trained networks (i.e., MobileNetV2, InceptionV3, and Xception) based on a quaternary-class MRI repository of glioma, meningioma, pituitary adenoma, and healthy cases. Adam optimization algorithm was used to train the model and the performance was evaluated using Gradient-weighted Class Activation Mapping visualization and quantitative outcomes of the metrics that included F1-score, classification accuracy, precision rates, sensitivity values, and confusion matrix.
Results: The architecture of MobileNetV2 presented better results since it attained a 95.7 percent accuracy in classification, and the precision, sensitivity and F1-score metrics of the model were harmonized. The attention-enhanced CNN had the same level of efficacy, with Grad-CAM heatmaps always showing pathologically significant areas, which confirmed the diagnostic reliability of these structures.
Conclusion: MobileNetV2 (pre-trained) architecture has outstanding performance in terms of MRI based neoplasm classification. Moreover, the combination of the attention mechanisms and Grad-CAM interpretability techniques can improve the accuracy and transparency of the diagnosis, and, therefore, they can be further adopted in the clinical diagnostic processes.
Correct classification of the tumor is crucial for the diagnosis and clinical management of neuro-oncology. Early tumor typing can help the clinician prioritize follow-up testing and plan for appropriate care. This study introduces an explainable brain tumor classification framework that combines a convolutional neural...
Zanbaq Hikmet Thanon, Yasir Dawood Salman, Shaymaa A. Kashmoola et al.· JOIV: International Journal...· 0 citations
A neurotumor multiclass classifier is presented, which has a probabilistic soft voting ensemble as the classification method, based on the transfer learning model using pre-trained Convolutional Neural Networks (CNNs), using VGG16, ResNet50 and fine-tuned EfficientNetB0 as the basic classifiers.
Anuj Gupta, Anita, Manish Gupta· international journal of eng...· 0 citations
A deep learning (DL)-based solution to automate brain tumor detection using a customized convolutional neural network (CNN) architecture tailored for multi-class classification of glioma, meningioma, pituitary tumors, and non-tumor cases is proposed.
Ghaida Alsharef, Raghad Asiri, Lama Alsultan et al.· Bulletin of Electrical Engin...· 0 citations
This research systematically benchmarks five CNN architectures (VGG19, DenseNet201, ResNet50, Inception-v3, and MobileNet) on balanced and naturally imbalanced MRI datasets, suggesting that VGG19 is particularly good at discriminative performance.
Tegar Anugrah Firdaus, B. Rais, Marcelinus Jonathan Salim et al.· 0 citations
Brain tumors are a leading cause of cancer-related mortality, and their early and accurate classification from magnetic resonance imaging (MRI) is critical for treatment planning. Manual interpretation is time-consuming and subject to inter-observer variability. This paper presents an automated four-class brain-tumor c...
Jun-Yao Zhang· Applied and Computational En...· 0 citations
A deep learning technique classifies MRI scans as pituitary tumor, glioma, meningioma, or no tumor, and provides explanations for its results, showing that the proposed model has strong discriminative power and high reliability for medical image classification tasks.