Motors constitute one critical part of industrial production and everyday life. The effective, timely and convenient diagnosis of motor faults is constantly required to ensure continuous and reliable operations. Infrared imaging technology, a non-invasive industrial fault diagnosis method, is usually applied to detect the equipment status in extreme environments. However, conventional Infrared thermal images inevitably show a large amount of noise interference, which affects the analysis results. In addition, each motor may only possess a small amount of fault data in practice, as collecting an infinite amount of motor data to train the diagnostic system is impossible. To overcome these problems, a novel automatic fault diagnosis system is proposed in this study. Data features are enhanced by a normalization module based on color bars first, as the same color in various infrared thermal images represent different temperatures. Then, the few-shot learning method is used to diagnose the faults of unseen electric motors. In the few-shot learning method, the minimum dataset size required to expand system universality is fifteen pieces, effectively solving the universality problem of artificial-to-natural data migration. The method saves a large amount of training data resources and the experimental training data collection. The accuracy of the fault diagnosis system achieved 98.9% on similar motor datasets and 91.8% on the dataset of motors that varied a lot from the training motor, which proves the high reliability and universality of the system.

Automatic and accurate segmentation and classification of esophageal lesions are two essential tasks to assist endoscopists in Upper Gastrointestinal Endoscopy. However, there is no intelligent system that can diagnose more lesion types, handle multiple tasks simultaneously, and be more accurate in clinical work. Therefore, we present an innovative Multi-Task deep learning architecture named Attention Gating Pyramid Vision Transformer (AGPVT), which provides a solution for the accurate classification and precise segmentation of lesion types and regions simultaneously. The proposed AGPVT combines the benefits of cutting-edge deep learning model designs with Multi-Task Learning (MTL) in order to advance the field. Furthermore, a patch-wise multi-head attention gating method alongside a hybrid design MTL decoder, is employed as the core driving architecture of the AGPVT. Comprehensive experiments are conducted on a multicenter dataset which contains esophageal cancer, Barrett's esophagus, esophageal protruded lesions, esophagitis, and normal esophagus. Experimental results show that the proposed AGPVT achieves a classification accuracy of 96.84%, an IoU score of 85.61%, and a Dice score of 90.75%, outperforming existing methods and demonstrating its effectiveness in this domain.

Pathologically, colorectal polyps can be classified as non-adenoma, which do not have a higher risk of developing colorectal cancer over time, and adenoma, which are thought to be precancerous lesions. The accurate identification and precise differentiation of colorectal polyps during colonoscopy play a crucial role in medical treatment. However, manual screening is prone to missed lesions and misinterpretations due to human error and the visual similarities between different types of polyps. Furthermore, the current process is expensive because it entails the pathological assessment of every polyp. With the development of deep learning technology, significant progress in target detection has made deep learning models a more accurate solution for medical imaging analysis. Therefore, to solve the problems of expensive biopsy costs, low identification accuracy, omission, and classification error caused by the similar appearance of non-adenoma and adenoma, a colorectal polyp detection algorithm based on an improved YOLOv5s equipped with a squeeze-and-excitation attention module was developed. The experiment showed that the performance metrics are higher than the original YOLOv5s, which shows the improved YOLOv5s have great potential to assist endoscopists in the decision-making process.