Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by mental abnormalities and motor dysfunction. Its early classification and prediction of clinical scores have been major concerns for researchers. Currently, multi-view data learning has become an essential research area due to the capacity of multiple views to provide complementary insights from various perspectives. However, the discontinuous distribution, data missing complexity, small sample size, and redundant features in multi-view datasets pose a substantial obstacle, and most existing multi-view learning methods are unable to handle these challenges effectively. In this study, we propose a novel incomplete multi-view data learning framework (IMVDL) via dynamic embedding and partiall2,1norm constraints for PD diagnosis. Specifically, multi-view dynamic embedding can adapt to any view missing scene, thereby linearly/nonlinearly mapping incomplete multi-view data to low-dimensional manifold spaces and generating complete multi-view data representations. The partiall2,1norm constraint can ignore larger feature weight values and performl2,1norm sparse on the remaining weights, thereby avoiding the sparse bias problem caused by larger weight values. An efficient iterative algorithm is derived to find the optimal solution of the IMVDL method. We conduct extensive experiments using multi-modal neuroimage data from the Parkinson's Progression Markers Initiative (PPMI) database. The results demonstrate that the IMVDL method is superior to other comparative methods. The source code for IMVDL is available at https://github.com/a610lab/IMVDL/.

Accurate prediction of fetal brain age from MRI is vital for assessing fetal development. However, clinical MRI data are often imbalanced across gestational ages, causing deep learning models to bias toward overrepresented samples and limiting generalization. Existing methods mainly adjust sample weights or smooth distributions but ignore ordinal structure consistency between feature and label spaces. To address this, we propose an adaptive multi-scale permutation entropy regularizer (AMPER) that enforces ordinal consistency between feature and label similarity sequences. AMPER introduces permutation entropy as a regularizer for deep regression, alleviating label imbalance bias and improving prediction reliability. It integrates (1) an adaptive scale selection mechanism guided by label entropy to ensure flexible regularization under varying data distributions, and (2) multi-scale permutation entropy to capture both local and global ordinal structures. Experiments show that AMPER achieves a mean absolute error of 0.793±0.024 weeks and an R2of 0.936±0.005, demonstrating its effectiveness for imbalanced regression and clinical potential for accurate fetal brain age estimation.

The classification of carotid plaques is crucial for assessing the risk of cardiovascular diseases. Recently, deep learning has emerged as a promising solution for automatic carotid plaque classification. However, its data-driven nature makes models prone to bias from noisy labels, often caused by inconsistent and unreliable labeling, which can significantly degrade model performance. To address this issue, this paper proposes a robust carotid plaque classification method (TMSS) designed to effectively learn from datasets with noisy labels using transition matrix and structural similarity regularization. The general framework of TMSS comprises an unsupervised constructive learning branch and a supervised classification branch, regularized by structural similarity to maximize the agreement of sample relationships between the feature space and the prediction space. The transition matrix is learned to further correct the noisy labels in the classifier branch. Evaluated on 1,270 carotid ultrasound images, experimental results demonstrate that TMSS achieves significant improvements in classification performance over state-of-the-art approaches. The proposed method enhances diagnostic accuracy by mitigating label noise and improving model robustness in carotid plaque classification, thereby facilitating early disease detection and enabling personalized treatment.

The segmentation and classification of carotid plaques in ultrasound images play important roles in the treatment of atherosclerosis and assessment for stroke risk. Although deep learning methods have been used for carotid plaque segmentation and classification, two-stage methods will increase the complexity of the overall analysis and the existing multi-task methods ignored the relationship between the segmentation and classification. These will lead to suboptimal performance as valuable information might not be fully leveraged across all tasks. Therefore, we propose a multi-task learning framework (RCCM-Net) for ultrasound carotid plaque segmentation and classification, which utilizes a region confidence module (RCM) and a sample category confidence module (CCM) to exploit the correlation between these two tasks. The RCM provides knowledge from the probability of plaque regions to the classification task, while the CCM is designed to learn the categorical sample weight for the segmentation task. A total of 1270 2D ultrasound images of carotid plaques were collected from Zhongnan Hospital (Wuhan, China) for our experiments. The results showed that the proposed method can improve both segmentation and classification performance compared to existing single-task networks (i.e., SegNet, Deeplabv3+, UNet++, EfficientNet, Res2Net, RepVGG, DPN) and multi-task algorithms (i.e., HRNet, MTANet), with an accuracy of 85.82% for classification and a Dice-similarity-coefficient of 84.92% for segmentation. In the ablation study, the results demonstrated that both the designed RCM and CCM were beneficial in improving the network's performance. Therefore, we believe that the proposed method could be useful for carotid plaque analysis in clinical practice.

Alzheimer’s disease is a neurodegenerative disorder that gradually impairs cognitive abilities. Early detection, diagnosis, and treatment are crucial for slowing the progression of the disease. In the diagnosis of Alzheimer’s disease, Graph Convolutional Networks (GCN) provide a powerful tool to enhance accuracy. However, the training of GCN faces challenges due to the tedious annotation process and limited data.To address this issue, we employ contrastive learning for pre-training GCN to improve classification performance under limited data conditions. Firstly, we augment graph data through singular value decomposition, preserving the brain’s primary topological structure and avoiding the loss of intrinsic semantic structure during augmentation. Secondly, we design a Soft-Clustering GCN to obtain more robust representations of brain data. Lastly, our framework clusters graphs with similar feature semantics into the same group and encourages clustering consistency between different augmentations of the same graph. In negative sampling, we select graphs from different groups as negative samples to ensure semantic differences between positive and negative samples. Experimental results demonstrate that our approach outperforms state-of-the-art methods on the Alzheimer’s disease dataset.

Parkinson’s disease (PD) is a common neurodegenerative disorder in the elderly population. The progressive symptoms of PD can have significant physical and economic implications for patients. Therefore, the development of a method to aid in the diagnosis and prediction of PD is crucial. However, medical neuroimaging data often have redundant features and high data dimensions, which can negatively impact algorithm accuracy. To solve this challenge, a supervised algorithm for feature selection is proposed for the early diagnosis and prediction of PD. Specifically, the proposed method incorporates adaptive learning during iterations, which allows adaptive updating of the similarity matrix and selection of informative features. Meanwhile, we introduce flexible mapping to address the limitation that linear mapping is too strict. To measure the effectiveness of the algorithm, we test it on the Parkinson’s Progression Markers Initiative (PPMI) public dataset. According to the outcomes of the experiment, the proposed method outperforms the competing feature selection methods and graph neural network approaches.

The analysis of plaque region in carotid ultrasound images is crucial for determining and assessing the harm-fulness of carotid plaques. Carotid ultrasound images provide both the location and status information of plaques, which can help diagnose carotid atherosclerosis. Despite this, the relationship between various plaque tasks has been disregarded in prior research, and due to the significant expense associated with manual image segmentation, there is a shortage of datasets that contain a substantial quantity of manually annotated plaque regions. In this paper, a semi-supervised learning algorithm is proposed to reduce reliance on annotated data, and due to the correlation between the plaque classification task and the plaque region semantic segmentation task, an auxiliary learning method named SAL-Net is proposed. The primary task of this model is supervised plaque classification, while the auxiliary task is a semi-supervised semantic segmentation task. The experiments are carried out on a carotid ultrasound image dataset, and the results show that SAL-net can effectively utilize the correlation between different tasks to improve the performance of the model.

Carotid plaque classification is of great significance for the diagnosis and treatment of carotid artery disease and the prediction of ischemic stroke. However, the ultrasound image structure of carotid plaques is complex, and manual labeling is time-consuming, which results in lots of unlabeled images. Semi-supervised methods can be well applied in this field by using unlabeled samples to improve model performance. Most of the existing semi-supervised methods are based on pseudo-labels and consistency regularization; however, these kinds of methods will cause error pseudo-labels during sample selection and an imbalanced distribution of pseudo-labels due to cognitive biases in model training. To solve these problems, We propose a method based on feature correction and pseudo-label balance correction, which utilizes a semi-supervised approach. The feature correction module uses the features of labeled data to correct the predicted probabilities of unlabeled data and improve the discrimination of pseudo-labels. The pseudo-label balance correction module can dynamically adjust the sample weight in the loss function according to the number of pseudo-labels to alleviate the problem of data imbalance. Evaluated on 1270 ultrasound carotid plaque images, our method achieved superior performance to other state-of-the-art methods (i.e., Mixmatch, Fixmatch, and Flexmatch) on 10%, 30%, and 50% labeled training data. Our method demonstrated accurate carotid plaque classification using a small number of labeled training datasets, potentially benefiting clinical practice and trials.

Measurement of total-plaque-area (TPA) is important for determining long term risk for stroke and monitoring carotid plaque progression. Since delineation of carotid plaques is required, a deep learning method can provide automatic plaque segmentations and TPA measurements; however, it requires large datasets and manual annotations for training with unknown performance on new datasets. A UNet++ ensemble algorithm was proposed to segment plaques from 2D carotid ultrasound images, trained on three small datasets (n = 33, 33, 34 subjects) and tested on 44 subjects from the SPARC dataset (n = 144, London, Canada). The ensemble was also trained on the entire SPARC dataset and tested with a different dataset (n = 497, Zhongnan Hospital, China). Algorithm and manual segmentations were compared using Dice-similarity-coefficient (DSC), and TPAs were compared using the difference (ΔTPA), Pearson correlation coefficient (r) and Bland-Altman analyses. Segmentation variability was determined using the intra-class correlation coefficient (ICC) and coefficient-of-variation (CoV). For 44 SPARC subjects, algorithm DSC was 83.3-85.7%, and algorithm TPAs were strongly correlated (r = 0.985-0.988; p <; 0.001) with manual results with marginal biases (0.73-6.75) mm$^2$ using the three training datasets. Algorithm ICC for TPAs (ICC = 0.996) was similar to intra- and inter-observer manual results (ICC = 0.977, 0.995). Algorithm CoV = 6.98% for plaque areas was smaller than the inter-observer manual CoV (7.54%). For the Zhongnan dataset, DSC was 88.6% algorithm and manual TPAs were strongly correlated (r = 0.972, p <; 0.001) with ΔTPA = -0.44±4.05 mm$^2$ and ICC = 0.985. The proposed algorithm trained on small datasets and segmented a different dataset without retraining with accuracy and precision that may be useful clinically and for research.

Vessel-wall-volume (VWV) is an important three-dimensional ultrasound (3DUS) metric used in the assessment of carotid plaque burden and monitoring changes in carotid atherosclerosis in response to medical treatment. To generate the VWV measurement, we proposed an approach that combined a voxel-based fully convolution network (Voxel-FCN) and a continuous max-flow module to automatically segment the carotid media-adventitia (MAB) and lumen-intima boundaries (LIB) from 3DUS images. Voxel-FCN includes an encoder consisting of a general 3D CNN and a 3D pyramid pooling module to extract spatial and contextual information, and a decoder using a concatenating module with an attention mechanism to fuse multi-level features extracted by the encoder. A continuous max-flow algorithm is used to improve the coarse segmentation provided by the Voxel-FCN. Using 1007 3DUS images, our approach yielded a Dice-similarity-coefficient (DSC) of 93.2±3.0% for the MAB in the common carotid artery (CCA), and 91.9±5.0% in the bifurcation by comparing algorithm and expert manual segmentations. We achieved a DSC of 89.5±6.7% and 89.3±6.8% for the LIB in the CCA and the bifurcation respectively. The mean errors between the algorithm-and manually-generated VWVs were 0.2±51.2 mm3for the CCA and -4.0±98.2 mm3for the bifurcation. The algorithm segmentation accuracy was comparable to intra-observer manual segmentation but our approach required less than 1s, which will not alter the clinical work-flow as 10s is required to image one side of the neck. Therefore, we believe that the proposed method could be used clinically for generating VWV to monitor progression and regression of carotid plaques.

Wireless capsule endoscopy (WCE) is a recently developed revolutionary medical technology which records the video of human's digestive tract noninvasively. However, reviewing a WCE video is a tired and time-consuming task for clinicians. Thus, WCE video automatic segmentation methods are emerging to reduce the review time for clinicians. In our previous work, a two-level WCE video segmentation approach has been proposed, which provides a novel approach to localize the boundaries more exactly and efficiently. However, it has an unsatisfactory performance in the small intestine/large intestine boundary detection. In this paper, we propose new features and an improved classifier to improve the previous two-level segmentation algorithm. In the rough level, color feature is utilized to draw a dissimilarity curve and an approximate boundary has been obtained. At the same time, training data for fine level can be directly labeled and collected between the two approximate boundaries of organs to overcome the difficulty of training data acquisition. In the fine level, a novel color uniform local binary pattern (CULBP) algorithm is proposed, which includes two kinds of patterns, color norm patterns and color angle patterns. The CULBP feature is more robust to variation of illumination and more discriminative for classification. Moreover, in order to elevate the performance of SVM classifier we proposed the Ada-SVM classifier which using RBFSVMs as component of Adaboost classifier. At last, an analysis of classification results of the Ada-SVM classifier is carried out to segment the WCE video into several meaningful parts, stomach, small intestine and large intestine. The experiments demonstrate a promising performance of the proposed method. The average precision and recall are as high as 91.37% and 88.50% in stomach/small intestine classification, 90.35% and 97.28% in small intestine/ large intestine classification.