The chronic neurodegenerative condition, which causes dementia and permanent cognitive loss in older persons, is known as Alzheimer’s Disease (AD). The early diagnosis of AD has recently benefited from the application of computer-aided technology. However, the diversity of AD neuroimaging and genetic data and the need for professional annotation of labels impact diagnostic performance. Taking gain of the multi-view data and relieving the problem triggered by way of the lack of labeling of part of the data, a novel Deep Learning (DL) model with Multi-view learning and Weakly-supervised learning based on a Transformer network is proposed, called MvWsT. The existence of consistency and complementarity of this data across different views is exploited to obtain a more powerful representation containing shared features and complementary features. At the same time, weakly-supervised learning is introduced to reduce the annotation of data, taking into account the particularity of the high cost of medical data annotation. The study utilizes Magnetic Resonance Imaging (MRI) to analyze neuroimaging in relation to AD, including two views in the axial view and sagittal view, with three Transformer models as baselines. Moreover, the proposed MvWsT method is validated by setting the unlabeled proportion and orthogonality constraints to complete the weakly supervised training. The results show the proposed MvWsT model has tremendous potential compared to the baselines with the single view.

As Machine Learning (ML) and Artificial Intelligence (AI) progress rapidly, the issue of ML model generalization has emerged as a critical concern for academics and practitioners alike. In practical scenarios, it is essential for models to sustain high performance when encountering varied and novel data distributions. Nevertheless, current domain generalization techniques have their shortcomings in tackling this challenge. The objective of this paper is to introduce a novel Meta-Learning approach, incorporating Fourier transform-based Data augmentation, called MLFD, for the purpose of domain generalization. Utilizing both data augmentation and a meta-learning architecture, this proposed technique empowers models to extend their generalization to multiple unseen target domains using just a single training domain. In contrast to other domain generalization methods, the method presented in this paper achieves comparable accuracy on the Digits-DG datasets, and demonstrates substantial improvements in terms of reducing model training time.

Underwater object detection technology is essential for maintaining marine ecological health and supporting economic development. However, the underwater environment poses significant challenges, including low contrast, small object sizes, and complex backgrounds. Existing generic object detectors often fail to identify these organisms effectively. This paper proposes a Joint Multi-scale channel attention and Multi-perception head Network (JMM-Net), a detection algorithm for underwater organisms. JMM-Net comprises three main components: Multi-Scale Channel Attention (MSCA)-based backbone network, Multi-Perception Parallel detection head (MPPhead), and lightweight GSconv-Path Aggregation Network (GS-PAN). MSCA is embedded into the backbone to enhance feature extraction for blurred and small-sized objects in low-quality environments by integrating local and global channel attention through multi-scale parallel sub-networks and cross-channel learning. MPPhead enhances the model’s classification and localization capabilities by leveraging scale, spatial, and task perception, thereby enhancing the detection of marine organisms in complex backgrounds. The adoption of GS-PAN over the traditional Path Aggregation Network (PAN) structure significantly reduces the model’s parameters and computational load, making it more suitable for deployment on edge devices. Extensive experiments on three public underwater datasets demonstrate that our method achieves excellent performance on underwater object detection at a lightweight cost.