Exudates can be regarded as one of the most prevalent clinical signs of diabetic retinopathy, and the detection of exudates has important clinical significance in diabetic retinopathy diagnosis. In this paper, a novel approach named superpixel multi-feature classification for the automatic detection of exudates is developed. First, an entire image is segmented into a series of superpixels considered as candidates. Then, a total of 20 features, including 19 multi-channel intensity features and a novel contextual feature, are proposed for characterizing each candidate. A supervised multi-variable classification algorithm is also introduced to distinguish the true exudates from the spurious candidates. Finally, a novel optic disc detection technique is designed to further improve the performance of classification accuracy. Extensive experiments are carried out on two publicly available online databases, DiaretDB1, and e-ophtha EX. Compared with other state-of-the-art approaches, the experimental results show the advantages and effectiveness of the proposed approach.

The temperature and pressure of electrical fused magnesia furnace (EFMF) are hard to obtain because the EFMF cannot be monitored effectively. Experts familiar with EFMF can determine its status by listening to its sound. In our work, we propose an intelligent architecture that can learn the experts' knowledge. Sound waveform is first transformed into the power spectrum density (PSD), and then the Laplacian score algorithm is used to select key frequencies. Next, a one-class support vector machine (SVM) is employed to find the minimum boundary of the submanifold of normal sound PSD. The distance of the sample to the classification hyperplane in kernel feature space is proposed to indicate the magnitude of faults. Lastly, the binary SVM is used to identify fault types. We have developed an instrument to acquire the sound of EFMF; experimental results using this data have shown the effectiveness of our proposed architecture. Additionally, the proposed architecture can be performed easily on the instrument for online monitoring.

The safe operation of aero-engines is crucial for ensuring flight safety, and effective fault detection methods are fundamental to achieving this objective. In this paper, we propose a novel approach that integrates an auto-encoder with long short-term memory (LSTM) networks and a self-attention mechanism for the anomaly detection of aero-engine time-series data. The dataset utilized in this study was simulated from real data and injected with fault information. A fault detection model is developed utilizing normal data samples for training and faulty data samples for testing. The LSTM auto-encoder processes the time-series data through an encoder–decoder architecture, extracting latent representations and reconstructing the original inputs. Furthermore, the self-attention mechanism captures long-range dependencies and significant features within the sequences, thereby enhancing the detection accuracy of the model. Comparative analyses with the traditional LSTM auto-encoder, as well as one-class support vector machines (OC-SVM) and isolation forests (IF), reveal that the experimental results substantiate the feasibility and effectiveness of the proposed method, highlighting its potential value in engineering applications.

Federated learning stands as a pivotal component in the construction of data infrastructure. It significantly fortifies the safety and reliability of data circulation links, facilitating credible sharing and openness among diverse subjects. The presence of missing data poses a pervasive and challenging issue in the implementation of federated learning. Current research on imputation missing values predominantly concentrates on centralized methods and horizontal federation scenarios. However, there is a notable absence of exploration in the context of vertical federated application scenarios. In this paper, the problem of missing imputation in vertical federated learning is investigated and a novel vertical federated k-nearest neighbors (KNN) imputation method is proposed. Extensive experiments are conducted using publicly available data sets to compare existing imputation methods, the results demonstrate the effectiveness and progress of our approach.

Red lesions can be regarded as one of the earliest lesions in diabetic retinopathy (DR) and automatic detection of red lesions plays a critical role in diabetic retinopathy diagnosis. In this paper, a novel superpixel Multichannel Multifeature (MCMF) classification approach is proposed for red lesion detection. In this paper, firstly, a new candidate extraction method based on superpixel is proposed. Then, these candidates are characterized by multichannel features, as well as the contextual feature. Next, FDA classifier is introduced to classify the red lesions among the candidates. Finally, a postprocessing technique based on multiscale blood vessels detection is modified for removing nonlesions appearing as red. Experiments on publicly available DiaretDB1 database are conducted to verify the effectiveness of our proposed method.