Verónica Silva Vasconcelos

Abstract
Gliomas are a common and aggressive kind of brain tumour that is difficult to diagnose due to their infiltrative development, variable clinical presentation, and complex behaviour, making them an important focus in neuro-oncology. Segmentation of brain tumour images is critical for improving diagnosis, prognosis, and treatment options. Manually segmenting brain tumours is time-consuming and challenging. Automatic segmentation algorithms can significantly improve the accuracy and efficiency of tumour identification, thus improving treatment planning and outcomes. Deep learning-based segmentation tumours have shown significant advances in the last few years. This study evaluates the impact of four denoising filters, namely median, Gaussian, anisotropic diffusion, and bilateral, on tumour detection and segmentation. The U-Net architecture is applied for the segmentation of 3064 contrast-enhanced magnetic resonance images from 233 patients diagnosed with meningiomas, gliomas, and pituitary tumours. The results of this work demonstrate that bilateral filtering yields superior outcomes, proving to be a robust and computationally efficient approach in brain tumour segmentation. This method reduces the processing time by 12 epochs, which in turn contributes to lowering greenhouse gas emissions by optimizing computational resources and minimizing energy consumption.

Abstract

Abstract
Breast cancer remains a significant global health concern. This study presents an image retrieval system to aid specialists in the analysis of mammogram images. The system employs individual classifiers for eight dimensions: laterality, view, breast density, BI-RADS classification, masses, calcifications, distortions, and asymmetries. Four pre-trained networks, ResNet50, VGG16, InceptionV3, and InceptionResNetV2, were used to train these classifiers. The retrieval model combines these classifiers through a weighted sum. Four weight assignment strategies were explored, ranging from equal weights to weights based on empirical, literature-based, and specialist-informed considerations. Results are illustrated using the INBreast database, which comprises 410 images. Besides the native annotations, ground truth to validate retrieval models had to be acquired. Classification accuracy is as high as 100% for some of the dimensions. Results also demonstrate the effectiveness of the proposed weighted-sum approach, with variations in weight assignments impacting model performance. © 2024 Owner/Author.

Abstract

Abstract