U-Net transfer learning backbones for lesions segmentation in breast ultrasound images

Abstract

<p>Breast ultrasound images are highly valuable for the early detection of breast cancer. However, the drawback of these images is low-quality resolution and the presence of speckle noise, which affects their interpretability and makes them radiologists’ expertise-dependent. As medical images, breast ultrasound datasets are scarce and imbalanced, and annotating them is tedious and time-consuming. Transfer learning, as a deep learning technique, can be used to overcome the dataset deficiency in available images. This paper presents the implementation of transfer learning U-Net backbones for the automatic segmentation of breast ultrasound lesions and implements a threshold selection mechanism to deliver optimal generalized segmentation results of breast tumors. The work uses the public breast ultrasound images (BUSI) dataset and implements ten state-of-theart candidate models as U-Net backbones. We have trained these models with a five-fold cross-validation technique on 630 images with benign and malignant cases. Five out of ten models showed good results, and the best U-Net backbone was found to be DenseNet121. It achieved an average Dice coefficient of 0.7370 and a sensitivity of 0.7255. The model’s robustness was also evaluated against normal cases, and the model accurately detected 72 out of 113 images, which is higher than the four best models.</p>

Automated DeepLabV3+ based model for left ventricle segmentation on short-axis late gadolinium enhancement-magnetic cardiac resonance imaging images

“…We adopted a dual-stage approach: Stage 1 employed MobileNetV2 [21], capitalizing on its lightweight design for swift feature extraction and LV chamber identification, minimizing computational burden for subsequent stages [22]. Stage 2 utilized ResNet50 [23], to extract finer details necessary for segmenting smaller scars [9], [24]- [26]. Both backbones were pre-trained on the ImageNet database and fine-tuned on their classification layers.…”

“…Recent studies have demonstrated the effective application of convolution neural networks (CNNs) in segmenting medical images across various fields such as breast cancer [9], [10] and lung segmentation [11]. Precise segmentation of the LV region, particularly areas containing scar tissue, provides a robust foundation for precise subsequent segmentation of myocardial scar tissue.…”

Revolutionizing Echocardiography: A Comparative Study of Advanced AI Models for Precise Left Ventricular Segmentation

“…2) U-Net: The U-Net structure has been intentionally designed to effectively tackle and overcome the diverse obstacles that arise when dealing with segmentation of medical images [24]. This model can be perceived as being split into two major components, specifically the contractive (encoding) path and the expansive (decoding) path.…”

Optimizing Ultrasound Image Classification Through Transfer Learning: Fine-Tuning Strategies and Classifier Impact on Pre-Trained Inner-Layers

“…The authors of the paper [3], investigated ten TL models as backbones for the U-Net [12] model for segmenting breast ultrasound images. The obtained results demonstrated the efficiency of pre-trained models in extracting relevant features for breast lesions segmentation.…”