Multi-label classification of data remains to be a challenging problem. Because of the complexity of the data, it is sometimes difficult to infer information about classes that are not mutually exclusive. For medical data, patients could have symptoms of multiple different diseases at the same time and it is important to develop tools that help to identify problems early. Intelligent health risk prediction models built with deep learning architectures offer a powerful tool for physicians to identify patterns in patient data that indicate risks associated with certain types of chronic diseases. Physical examination records of 110,300 anonymous patients were used to predict diabetes, hypertension, fatty liver, a combination of these three chronic diseases, and the absence of disease (8 classes in total). The dataset was split into training (90%) and testing (10%) sub-datasets. Ten-fold cross validation was used to evaluate prediction accuracy with metrics such as precision, recall, and F-score. Deep Learning (DL) architectures were compared with standard and state-of-the-art multi-label classification methods. Preliminary results suggest that Deep Neural Networks (DNN), a DL architecture, when applied to multi-label classification of chronic diseases, produced accuracy that was comparable to that of common methods such as Support Vector Machines. We have implemented DNNs to handle both problem transformation and algorithm adaption type multi-label methods and compare both to see which is preferable. Deep Learning architectures have the potential of inferring more information about the patterns of physical examination data than common classification methods. The advanced techniques of Deep Learning can be used to identify the significance of different features from physical examination data as well as to learn the contributions of each feature that impact a patient’s risk for chronic diseases. However, accurate prediction of chronic disease risks remains a challenging problem that warrants further studies.

Chronic diseases are one of the biggest threats to human life. It is clinically significant to predict the chronic disease prior to diagnosis time and take effective therapy as early as possible. In this work, we use problem transform methods to convert the chronic diseases prediction into a multi-label classification problem and propose a novel convolutional neural network (CNN) architecture named GroupNet to solve the multi-label chronic disease classification problem. Binary Relevance (BR) and Label Powerset (LP) methods are adopted to transform multiple chronic disease labels. We present the correlated loss as the loss function used in the GroupNet, which integrates the correlation coefficient between different diseases. The experiments are conducted on the physical examination datasets collected from a local medical center. In the experiments, we compare GroupNet with other methods and models. GroupNet outperforms others and achieves the best accuracy of 81.13%.

Salient and small lesions (e.g., microaneurysms on fundus) both play significant roles in real-world disease diagnosis under medical image examinations. Although deep neural networks (DNNs) have achieved promising medical image classification performance, they often have limitations in capturing both salient and small lesion information, restricting performance improvement in imbalanced medical image classification. Recently, with the advent of DNN-based style transfer in medical image generation, the roles of clinical styles have attracted great interest, as they are crucial indicators of lesions. Motivated by this observation, we propose a novel Adaptive Dual-Axis Style-based Recalibration (ADSR) module, leveraging the potential of clinical styles to guide DNNs in effectively learning salient and small lesion information from a dual-axis perspective. ADSR first emphasizes salient lesion information via global style-based adaptation, then captures small lesion information with pixel-wise style-based fusion. We construct an ADSR-Net for imbalanced medical image classification by stacking multiple ADSR modules. Additionally, DNNs typically adopt cross-entropy loss for parameter optimization, which ignores the impacts of class-wise predicted probability distributions. To address this, we introduce a new Class-wise Statistics Loss (CWS) combined with CE to further boost imbalanced medical image classification results. Extensive experiments on five imbalanced medical image datasets demonstrate not only the superiority of ADSR-Net and CWS over state-of-the-art (SOTA) methods but also their improved confidence calibration results. For example, ADSR-Net with the proposed loss significantly outperforms CABNet50 by 21.39% and 27.82% in F1 and B-ACC while reducing 3.31% and 4.57% in ECE and BS on ISIC2018.

Stroke is one of the leading causes of disability and incidence. For the treatment of prognosis of stroke patients, Motor imagery (MI) as a novel experimental paradigm, clinically it is effective because MI based Brain-Computer interface system can promote rehabilitation of stroke patients. There is being a hot and challenging topic to recognize multi-class motor imagery action classification accurately based on electroencephalograph (EEG) signals. In this work, we propose a novel framework named MRC-MLP. Multiple Riemannian covariance is used for EEG feature extraction. We make a multi-scale spectral division to filter EEG signals. They consist of different frequency bandwidths name sub-band. We concatenate and vectorize features extracted by Riemannian covariance on each sub-band. We design a fully connected MLP model with an improved loss function for motor imagery EEG classification. Furthermore, our proposed method MRC-MLP outperforms state-of-the-art methods and achieves approximately mean accuracy with 76%.

It is important to identify and prevent disease risk as early as possible through regular physical examinations. We formulate the disease risk prediction into a multilabel classification problem. A novel Ensemble Label Power-set Pruned datasets Joint Decomposition (ELPPJD) method is proposed in this work. First, we transform the multilabel classification into a multiclass classification. Then, we propose the pruned datasets and joint decomposition methods to deal with the imbalance learning problem. Two strategies size balanced (SB) and label similarity (LS) are designed to decompose the training dataset. In the experiments, the dataset is from the real physical examination records. We contrast the performance of the ELPPJD method with two different decomposition strategies. Moreover, the comparison between ELPPJD and the classic multilabel classification methods RAkEL and HOMER is carried out. The experimental results show that the ELPPJD method with label similarity strategy has outstanding performance.

Ultrasonography is widely used in the clinical diagnosis of thyroid nodules. Ultrasound images of thyroid nodules have different appearances, interior features, and blurred borders that are difficult for a physician to diagnose into malignant or benign types merely through visual recognition. The development of artificial intelligence, especially deep learning, has led to great advances in the field of medical image diagnosis. However, there are some challenges to achieve precision and efficiency in the recognition of thyroid nodules. In this work, we propose a deep learning architecture, you only look once v3 dense multireceptive fields convolutional neural network (YOLOv3-DMRF), based on YOLOv3. It comprises a DMRF-CNN and multiscale detection layers. In DMRF-CNN, we integrate dilated convolution with different dilation rates to continue passing the edge and the texture features to deeper layers. Two different scale detection layers are deployed to recognize the different sizes of the thyroid nodules. We used two datasets to train and evaluate the YOLOv3-DMRF during the experiments. One dataset includes 699 original ultrasound images of thyroid nodules collected from a local health physical center. We obtained 10,485 images after data augmentation. Another dataset is an open-access dataset that includes ultrasound images of 111 malignant and 41 benign thyroid nodules. Average precision (AP) and mean average precision (mAP) are used as the metrics for quantitative and qualitative evaluations. We compared the proposed YOLOv3-DMRF with some state-of-the-art deep learning networks. The experimental results show that YOLOv3-DMRF outperforms others on mAP and detection time on both the datasets. Specifically, the values of mAP and detection time were 90.05 and 95.23% and 3.7 and 2.2 s, respectively, on the two test datasets. Experimental results demonstrate that the proposed YOLOv3-DMRF is efficient for detection and recognition of thyroid nodules for ultrasound images.