RESUMEN

BACKGROUND: In the last decade, long-tail learning has become a popular research focus in deep learning applications in medicine. However, no scientometric reports have provided a systematic overview of this scientific field. We utilized bibliometric techniques to identify and analyze the literature on long-tailed learning in deep learning applications in medicine and investigate research trends, core authors, and core journals. We expanded our understanding of the primary components and principal methodologies of long-tail learning research in the medical field. METHODS: Web of Science was utilized to collect all articles on long-tailed learning in medicine published until December 2023. The suitability of all retrieved titles and abstracts was evaluated. For bibliometric analysis, all numerical data were extracted. CiteSpace was used to create clustered and visual knowledge graphs based on keywords. RESULTS: A total of 579 articles met the evaluation criteria. Over the last decade, the annual number of publications and citation frequency both showed significant growth, following a power-law and exponential trend, respectively. Noteworthy contributors to this field include Husanbir Singh Pannu, Fadi Thabtah, and Talha Mahboob Alam, while leading journals such as IEEE ACCESS, COMPUTERS IN BIOLOGY AND MEDICINE, IEEE TRANSACTIONS ON MEDICAL IMAGING, and COMPUTERIZED MEDICAL IMAGING AND GRAPHICS have emerged as pivotal platforms for disseminating research in this area. The core of long-tailed learning research within the medical domain is encapsulated in six principal themes: deep learning for imbalanced data, model optimization, neural networks in image analysis, data imbalance in health records, CNN in diagnostics and risk assessment, and genetic information in disease mechanisms. CONCLUSION: This study summarizes recent advancements in applying long-tail learning to deep learning in medicine through bibliometric analysis and visual knowledge graphs. It explains new trends, sources, core authors, journals, and research hotspots. Although this field has shown great promise in medical deep learning research, our findings will provide pertinent and valuable insights for future research and clinical practice.

Asunto(s)

Asunto(s)

RESUMEN

Generative adversarial networks (GANs) and their variants as an effective method for generating visually appealing images have shown great potential in different medical imaging applications during past decades. However, some issues remain insufficiently investigated: many models still suffer from model collapse, vanishing gradients, and convergence failure. Considering the fact that medical images differ from typical RGB images in terms of complexity and dimensionality, we propose an adaptive generative adversarial network, namely MedGAN, to mitigate these issues. Specifically, we first use Wasserstein loss as a convergence metric to measure the convergence degree of the generator and the discriminator. Then, we adaptively train MedGAN based on this metric. Finally, we generate medical images based on MedGAN and use them to build few-shot medical data learning models for disease classification and lesion localization. On demodicosis, blister, molluscum, and parakeratosis datasets, our experimental results verify the advantages of MedGAN in model convergence, training speed, and visual quality of generated samples. We believe this approach can be generalized to other medical applications and contribute to radiologists' efforts for disease diagnosis. The source code can be downloaded at https://github.com/geyao-c/MedGAN.

Asunto(s)

RESUMEN

Medical images are an important basis for doctors to diagnose diseases, but some medical images have low resolution due to hardware technology and cost constraints. Super-resolution technology can reconstruct low-resolution medical images into high-resolution images and enhance the quality of low-resolution images, thus assisting doctors in diagnosing diseases. However, traditional super-resolution methods mainly learn the mapping relationships among modal pixels from low resolution to high resolution, lacking the learning of high-level semantic features, resulting in a lack of understanding and utilization of semantic information, such as reconstructed objects, object attributes, and spatial relationships between two objects. In this paper, we propose a medical image super-resolution method based on semantic perception transfer learning. First, we propose a novel semantic perception super-resolution method that empowers super-resolution models to perceive high-level semantics by transferring features of the image description generation network in natural language processing. Second, we construct a semantic feature extraction network and an image description generation network and comprehensively utilized image and text modal data to learn transferable, high-level semantic features. Third, we train an end-to-end, semantic perception super-resolution model by fusing dynamic perceptual convolution, a semantic extraction network, and distillation polarization self-attention. Experiments show that semantic perception transfer learning can effectively improve the quality of super-resolution reconstruction.

RESUMEN

Skin diseases are one of the most common ailments affecting humans. Artificial intelligence based on deep learning can significantly improve the efficiency of identifying skin disorders and alleviate the scarcity of medical resources. However, the distribution of background information in dermatological datasets is imbalanced, causing generalized deep learning models to perform poorly in skin disease classification. We propose a deep learning schema that combines data preprocessing, data augmentation, and residual networks to study the influence of color-based background selection on a deep model's capacity to learn foreground lesion subject attributes in a skin disease classification problem. First, clinical photographs are annotated by dermatologists, and then the original background information is masked with unique colors to generate several subsets with distinct background colors. Sample-balanced training and test sets are generated using random over/undersampling and data augmentation techniques. Finally, the deep learning networks are independently trained on diverse subsets of backdrop colors to compare the performance of classifiers based on different background information. Extensive experiments demonstrate that color-based background information significantly affects the classification of skin diseases and that classifiers trained on the green subset achieve state-of-the-art performance for classifying black and red skin lesions.

Asunto(s)

RESUMEN

Computer-aided diagnosis (CAD) has always been an important research topic for applying artificial intelligence in smart healthcare. Sufficient medical data are one of the most critical factors in CAD research. However, medical data are usually obtained in chronological order and cannot be collected all at once, which poses difficulties for the application of deep learning technology in the medical field. The traditional batch learning method consumes considerable time and space resources for real-time medical data, and the incremental learning method often leads to catastrophic forgetting. To solve these problems, we propose a real-time medical data processing method based on federated learning. We divide the process into the model stage and the exemplar stage. In the model stage, we use the federated learning method to fuse the old and new models to mitigate the catastrophic forgetting problem of the new model. In the exemplar stage, we use the most representative exemplars selected from the old data to help the new model review the old knowledge, which further mitigates the catastrophic forgetting problem of the new model. We use this method to conduct experiments on a simulated medical real-time data stream. The experimental results show that our method can learn a disease diagnosis model from a continuous medical real-time data stream. As the amount of data increases, the performance of the disease diagnosis model continues to improve, and the catastrophic forgetting problem has been effectively mitigated. Compared with the traditional batch learning method, our method can significantly save time and space resources.

RESUMEN

There are two key requirements for medical lesion image super-resolution reconstruction in intelligent healthcare systems: clarity and reality. Because only clear and real super-resolution medical images can effectively help doctors observe the lesions of the disease. The existing super-resolution methods based on pixel space optimization often lack high-frequency details which result in blurred detail features and unclear visual perception. Also, the super-resolution methods based on feature space optimization usually have artifacts or structural deformation in the generated image. This paper proposes a novel pyramidal feature multi-distillation network for super-resolution reconstruction of medical images in intelligent healthcare systems. Firstly, we design a multi-distillation block that combines pyramidal convolution and shallow residual block. Secondly, we construct a two-branch super-resolution network to optimize the visual perception quality of the super-resolution branch by fusing the information of the gradient map branch. Finally, we combine contextual loss and L1 loss in the gradient map branch to optimize the quality of visual perception and design the information entropy contrast-aware channel attention to give different weights to the feature map. Besides, we use an arbitrary scale upsampler to achieve super-resolution reconstruction at any scale factor. The experimental results show that the proposed super-resolution reconstruction method achieves superior performance compared to other methods in this work.

RESUMEN

Elderly patients with coronavirus disease 2019 (COVID-19) are more likely to develop severe or critical pneumonia, with a high fatality rate. To date, there is no model to predict the severity of COVID-19 in elderly patients. In this study, patients who maintained a non-severe condition and patients who progressed to severe or critical COVID-19 during hospitalization were assigned to the non-severe and severe groups, respectively. Based on the admission data of these two groups in the training cohort, albumin (odds ratio [OR] = 0.871, 95% confidence interval [CI]: 0.809 - 0.937, P < 0.001), d-dimer (OR = 1.289, 95% CI: 1.042 - 1.594, P = 0.019) and onset to hospitalization time (OR = 0.935, 95% CI: 0.895 - 0.977, P = 0.003) were identified as significant predictors for the severity of COVID-19 in elderly patients. By combining these predictors, an effective risk nomogram was established for accurate individualized assessment of the severity of COVID-19 in elderly patients. The concordance index of the nomogram was 0.800 in the training cohort and 0.774 in the validation cohort. The calibration curve demonstrated excellent consistency between the prediction of our nomogram and the observed curve. Decision curve analysis further showed that our nomogram conferred significantly high clinical net benefit. Collectively, our nomogram will facilitate early appropriate supportive care and better use of medical resources and finally reduce the poor outcomes of elderly COVID-19 patients.

Asunto(s)

RESUMEN

Secure data aggregation (SDA) schemes are widely used in distributed applications, such as mobile sensor networks, to reduce communication cost, prolong the network life cycle and provide security. However, most SDA are only suited for a single type of statistics (i.e., summation-based or comparison-based statistics) and are not applicable to obtaining multiple statistic results. Most SDA are also inefficient for dynamic networks. This paper presents multi-functional secure data aggregation (MFSDA), in which the mapping step and coding step are introduced to provide value-preserving and order-preserving and, later, to enable arbitrary statistics support in the same query. MFSDA is suited for dynamic networks because these active nodes can be counted directly from aggregation data. The proposed scheme is tolerant to many types of attacks. The network load of the proposed scheme is balanced, and no significant bottleneck exists. The MFSDA includes two versions: MFSDA-I and MFSDA-II. The first one can obtain accurate results, while the second one is a more generalized version that can significantly reduce network traffic at the expense of less accuracy loss.

RESUMEN

Searching useful information from unstructured medical multimedia data has been a difficult problem in information retrieval. This paper reports an effective semantic medical multimedia retrieval approach which can reflect the users' query intent. Firstly, semantic annotations will be given to the multimedia documents in the medical multimedia database. Secondly, the ontology that represented semantic information will be hidden in the head of the multimedia documents. The main innovations of this approach are cross-type retrieval support and semantic information preservation. Experimental results indicate a good precision and efficiency of our approach for medical multimedia retrieval in comparison with some traditional approaches.

Asunto(s)