Breast ultrasound (BUS) imaging is commonly used in the early detection of breast cancer as a portable, valuable, and widely available diagnosis tool. Automated BUS image classification segmentation can assist radiologists making accurate fast decisions. Recent studies illustrate that tumor, peritumoral, background regions images provide valuable information for or classification. However, few have investigated influence these three on multi-task learning. In this study, we propose an...

Ultrasound imaging is one of the most commonly used diagnostic tools to detect and classify abnormalities women breast. Automatic ultrasound image segmentation provides radiologists a second opinion increase diagnosis accuracy. Deep neural networks have recently been employed achieve better results than conventional approaches. In this paper, we propose novel deep learning architecture, Multi-Scale Self-Attention Network (MSSA-Net), which can be trained on small datasets explore...

Breast Ultrasound (BUS) imaging is an essential tool for the early detection of breast cancer. The Imaging Reporting and Data System (BI-RADS) in BUS images helps standardize interpretation reporting process by categorizing tumors into multiple classes, which enables radiologists to make more accurate diagnoses treatment plans. However, most existing classification methods distinguish only between benign malignant categories. In addition, features extracted classic convolutional neural...

Computer-aided diagnosis (CAD) can help doctors in diagnosing breast cancer. Breast ultrasound (BUS) imaging is harmless, effective, portable, and the most popular modality for cancer detection/diagnosis. Many researchers work on improving performance of CAD systems. However, there are two main shortcomings: (1) Most existing methods based prerequisites that one only tumor image. (2) The results depend datasets, i.e., an algorithm using different datasets may obtain performances. It implies...

Breast cancer is a great threat to women’s health. Automatic analysis of UltraSound (BUS) images can help radiologists make more accurate and efficient diagnoses breast cancer. We propose Multi-Task Learning Network with Context-Oriented Self-Attention (MTL-COSA) module automatically simultaneously segment tumors classify them as benign or malignant. The COSA incorporates prior medical knowledge guide the network learn contextual relationships for better feature representations in BUS...

The study explores the transformative capabilities of Transformers and Large Language Models (LLMs) in early detection Acute Lymphoblastic Leukaemia (ALL). researchers benchmark Vision with Deformable Attention (DAT) Hierarchical (Swin) against established Convolutional Neural Networks (CNNs) like ResNet-50 VGG-16. findings reveal that transformer models exhibit remarkable accuracy identifying ALL from original images, demonstrating efficiency image analysis without necessitating...

Despite significant advancements in Artificial Intelligence (AI) and Large Language Models (LLMs), detecting mitigating bias remains a critical challenge, particularly on social media platforms like X (formerly Twitter), to address the prevalent cyberbullying these platforms. This research investigates effectiveness of leading LLMs generating synthetic biased data evaluates proficiency transformer AI models within both authentic contexts. The study involves semantic analysis feature...

Convolutional neural networks (CNNs) are widely used in medical image analysis, especially for breast ultrasound (BUS) segmentation. Automatically encoding deep features is one of the most important reasons leading to success convolutional networks. There a lot studies on obtaining better features; how-ever, they do not discuss higher-order information features. In this research, we propose novel operator, shape-adaptive which can select pixels calculating convolution rather than Euclidean...

In an era where artificial intelligence (AI) bridges crucial communication gaps, this study extends AI’s utility to American and Taiwan Sign Language (ASL TSL) communities through advanced models like the hierarchical vision transformer with shifted windows (Swin). This research evaluates Swin’s adaptability across sign languages, aiming for a universal platform unvoiced. Utilizing deep learning technologies, it has developed prototypes ASL-to-English translation, supported by educational...

Chest X-ray is widely used to diagnose lung diseases. Due the demand for accelerating analysis and interpretation reduce workload of radiologists, there has been a growing interest in building automated systems chest abnormality localization. However, fully supervised methods usually require well-trained radiologists annotate bounding boxes manually, which labor-intensive time-consuming. As result, weakly localization gaining increasing attention because it only requires image-level...

In the era of Artificial Intelligence (AI), comprehending and responding to non-verbal communication is increasingly vital. This research extends AI's reach in bridging gaps, notably benefiting American Sign Language (ASL) Taiwan (TSL) communities. It focuses on employing various AI models, especially Hierarchical Vision Transformer with Shifted Windows (Swin), for recognizing diverse sign language datasets. The study assesses Swin architecture's adaptability different...

Despite significant advancements in Artificial Intelligence (AI) and Large Language Models (LLMs), detecting mitigating bias remains a critical challenge, particularly within social media platforms like X (formerly Twitter) addressing cyberbullying present on them. This research investigates the effectiveness of leading LLMs generating synthetic biased data evaluates proficiency Transformer AI models both authentic contexts. The study involves semantic analysis feature engineering dataset...

Osteoarthritis (OA) is one of the major health issues among elderly population. MRI most popular technology to observe and evaluate progress OA course. However, extreme labor cost analysis makes process inefficient expensive. Also, due human error subjective nature, inter- intra-observer variability rather high. Computer-aided knee segmentation currently an active research field because it can alleviate doctors radiologists from time consuming tedious job, improve diagnosis performance which...

Medical image semantic segmentation is essential in computer-aided diagnosis systems. It can separate tissues and lesions the provide valuable information to radiologists doctors. The breast ultrasound (BUS) images have advantages: no radiation, low cost, portable, etc. However, there are two unfavorable characteristics: (1) dataset size often small due difficulty obtaining ground truths, (2) BUS usually poor quality. Trustworthy urgent cancer systems, especially for fully understanding...

Automatic tumor segmentation of breast ultrasound (BUS) image is quite challenging due to the complicated anatomic structure and poor quality. Most approaches achieve good performance on BUS images collected in controlled settings; however, degrades greatly with from different sources. Tumor saliency estimation (TSE) has attracted increasing attention solve problem by modeling radiologists' mechanism. In this paper, we propose a novel hybrid framework for TSE, which integrates both...

Automatic breast ultrasound image (BUS) segmentation is still a challenging task due to poor quality and inherent speckle noise. In this paper, we propose novel multi-scale fuzzy generative adversarial network (MSF-GAN) for segmentation. The proposed MSF-GAN consists of two networks: generate maps input BUS images, discriminative that employs (MSF) entropy module discrimination. major contribution paper applying logic in the which can distinguish uncertainty groundtruth forces achieve better...

Breast cancer is one of the most serious disease affecting women's health. Due to low cost, portable, no radiation, and high efficiency, breast ultrasound (BUS) imaging popular approach for diagnosing early cancer. However, images are resolution poor quality. Thus, developing accurate detection system a challenging task. In this paper, we propose fully automatic segmentation algorithm consisting two parts: fuzzy convolutional network accurately fine-tuning post-processing based on anatomy...