Microaneurysm is the initial symptom of diabetic retinopathy. Eliminating this lesion can effectively prevent diabetic retinopathy in the early stage. However, due to the complex retinal structure and the different brightness and contrast of fundus image because of different factors such as patients, environment and acquisition equipment, the existing detection algorithms are difficult to achieve the accurate detection and location of the lesion. Therefore, an improved detection algorithm of you only look once (YOLO) v4 with Squeeze-and-Excitation networks (SENet) embedded was proposed. Firstly, an improved and fast fuzzy c-means clustering algorithm was used to optimize the anchor parameters of the target samples to improve the matching degree between the anchors and the feature graphs; Then, the SENet attention module was embedded in the backbone network to enhance the key information of the image and suppress the background information of the image, so as to improve the confidence of microaneurysms; In addition, an spatial pyramid pooling was added to the network neck to enhance the acceptance domain of the output characteristics of the backbone network, so as to help separate important context information; Finally, the model was verified on the Kaggle diabetic retinopathy dataset and compared with other methods. The experimental results showed that compared with other YOLOv4 network models with various structures, the improved YOLOv4 network model could significantly improve the automatic detection results such as F-score which increased by 12.68%; Compared with other network models and methods, the automatic detection accuracy of the improved YOLOv4 network model with SENet embedded was obviously better, and accurate positioning could be realized. Therefore, the proposed YOLOv4 algorithm with SENet embedded has better performance, and can accurately and effectively detect and locate microaneurysms in fundus images.
Network meta-analysis may be performed by fitting multivariate meta-analysis models with Stata software mvmeta command; however, there are various challenges such as preprocessing the data, parameterising the model, and making good graphical displays of results. A suite of Stata programs, network, may meet these challenges. In this article, we introduce how to use the network commands to implement network meta-analysis by the example of continuous data.
Electrocardiogram (ECG) is a noninvasive, inexpensive, and convenient test for diagnosing cardiovascular diseases and assessing the risk of cardiovascular events. Although there are clear standardized operations and procedures for ECG examination, the interpretation of ECG by even trained physicians can be biased due to differences in diagnostic experience. In recent years, artificial intelligence has become a powerful tool to automatically analyze medical data by building deep neural network models, and has been widely used in the field of medical image diagnosis such as CT, MRI, ultrasound and ECG. This article mainly introduces the application progress of deep neural network models in ECG diagnosis and prediction of cardiovascular diseases, and discusses its limitations and application prospects.
In the extraction of fetal electrocardiogram (ECG) signal, due to the unicity of the scale of the U-Net same-level convolution encoder, the size and shape difference of the ECG characteristic wave between mother and fetus are ignored, and the time information of ECG signals is not used in the threshold learning process of the encoder’s residual shrinkage module. In this paper, a method of extracting fetal ECG signal based on multi-scale residual shrinkage U-Net model is proposed. First, the Inception and time domain attention were introduced into the residual shrinkage module to enhance the multi-scale feature extraction ability of the same level convolution encoder and the utilization of the time domain information of fetal ECG signal. In order to maintain more local details of ECG waveform, the maximum pooling in U-Net was replaced by Softpool. Finally, the decoder composed of the residual module and up-sampling gradually generated fetal ECG signals. In this paper, clinical ECG signals were used for experiments. The final results showed that compared with other fetal ECG extraction algorithms, the method proposed in this paper could extract clearer fetal ECG signals. The sensitivity, positive predictive value, and F1 scores in the 2013 competition data set reached 93.33%, 99.36%, and 96.09%, respectively, indicating that this method can effectively extract fetal ECG signals and has certain application values for perinatal fetal health monitoring.
The electroencephalogram (EEG) signal is a general reflection of the neurophysiological activity of the brain, which has the advantages of being safe, efficient, real-time and dynamic. With the development and advancement of machine learning research, automatic diagnosis of Alzheimer’s diseases based on deep learning is becoming a research hotspot. Started from feedforward neural networks, this paper compared and analysed the structural properties of neural network models such as recurrent neural networks, convolutional neural networks and deep belief networks and their performance in the diagnosis of Alzheimer’s disease. It also discussed the possible challenges and research trends of this research in the future, expecting to provide a valuable reference for the clinical application of neural networks in the EEG diagnosis of Alzheimer’s disease.
Background music has been increasingly affecting people’s lives. The research on the influence of background music on working memory has become a hot topic in brain science. In this paper, an improved electroencephalography (EEG) experiment based on n-back paradigm was designed. Fifteen university students without musical training were randomly selected to participate in the experiment, and their behavioral data and the EEG data were collected synchronously in order to explore the influence of different types of background music on spatial positioning cognition working memory. The exact low-resolution brain tomography algorithm (eLORETA) was applied to localize the EEG sources and the cross-correlation method was used to construct the cortical brain function networks based on the EEG source signals. Then the characteristics of the networks under different conditions were analyzed and compared to study the effects of background music on people’s working memory. The results showed that the difference of peak periods after stimulated by different types of background music were mainly distributed in the signals of occipital lobe and temporal lobe (P < 0.05). The analysis results showed that the brain connectivity under the condition with background music were stronger than those under the condition without music. The connectivities in the right occipital and temporal lobes under the condition of rock music were significantly higher than those under the condition of classical music. The node degrees, the betweenness centrality and the clustering coefficients under the condition without music were lower than those under the condition with background music. The node degrees and clustering coefficients under the condition of classical music were lower than those under the condition of rock music. It indicates that music stimulation increases the brain activity and has an impact on the working memory, and the effect of rock music is more remarkable than that of classical music. The behavioral data showed that the response accuracy in the state of no music, classical music and rock music were 86.09% ± 0.090%, 80.96% ± 0.960% and 79.36% ± 0.360%, respectively. We conclude that background music has a negative impact on the working memory, for it takes up the cognitive resources and reduces the cognitive ability of spatial location.
Objective To develop a deep learning system for CT images to assist in the diagnosis of thoracolumbar fractures and analyze the feasibility of its clinical application. Methods Collected from West China Hospital of Sichuan University from January 2019 to March 2020, a total of 1256 CT images of thoracolumbar fractures were annotated with a unified standard through the Imaging LabelImg system. All CT images were classified according to the AO Spine thoracolumbar spine injury classification. The deep learning system in diagnosing ABC fracture types was optimized using 1039 CT images for training and validation, of which 1004 were used as the training set and 35 as the validation set; the rest 217 CT images were used as the test set to compare the deep learning system with the clinician’s diagnosis. The deep learning system in subtyping A was optimized using 581 CT images for training and validation, of which 556 were used as the training set and 25 as the validation set; the rest 104 CT images were used as the test set to compare the deep learning system with the clinician’s diagnosis. Results The accuracy and Kappa coefficient of the deep learning system in diagnosing ABC fracture types were 89.4% and 0.849 (P<0.001), respectively. The accuracy and Kappa coefficient of subtyping A were 87.5% and 0.817 (P<0.001), respectively. Conclusions The classification accuracy of the deep learning system for thoracolumbar fractures is high. This approach can be used to assist in the intelligent diagnosis of CT images of thoracolumbar fractures and improve the current manual and complex diagnostic process.
Considering the abundant vascular anastomotic networks in the deep fascia of the posterior calf, three kinds of distally based facial flap containing deep fascial vascular network were applied clinically. They were: 1. posterolateral distally based island fascial flap which could be used to repair the skin defect of heel, dorsum of foot and lateral-distal part of leg; 2. posteromedial distally based island fascial flap which could be used to repair the skin defect of heel, medial malleolus and medial-distal part of leg and 3. posterolateral malleolar distally based fascial flap which could be used to repair the skin defect of heel and lateral malleolus. Eighteen cases with soft tissue defects around the distal calf were treated, the area of skin defect ranged from 4 cm x 3 cm to 13 cm x 6 cm. All the flaps were survived completely after operation with an average of follow-up for 15 months (ranged from 6 months to 2 years). So the advantages of these flaps were as follows: the blood supply was reliable, preparation of the flap was easy and the major arteries of the calf needed not be sacrificed; the flap had a long and rotatable pedicle so that they would basically satisfy the need to repair skin defect of lower leg, dorsum of foot, heel and malleolus and the resistance of the flap to pressure and wear was better. However, the injury to the superficial sural nerve was the shortcoming.
Deep learning method can be used to automatically analyze electrocardiogram (ECG) data and rapidly implement arrhythmia classification, which provides significant clinical value for the early screening of arrhythmias. How to select arrhythmia features effectively under limited abnormal sample supervision is an urgent issue to address. This paper proposed an arrhythmia classification algorithm based on an adaptive multi-feature fusion network. The algorithm extracted RR interval features from ECG signals, employed one-dimensional convolutional neural network (1D-CNN) to extract time-domain deep features, employed Mel frequency cepstral coefficients (MFCC) and two-dimensional convolutional neural network (2D-CNN) to extract frequency-domain deep features. The features were fused using adaptive weighting strategy for arrhythmia classification. The paper used the arrhythmia database jointly developed by the Massachusetts Institute of Technology and Beth Israel Hospital (MIT-BIH) and evaluated the algorithm under the inter-patient paradigm. Experimental results demonstrated that the proposed algorithm achieved an average precision of 75.2%, an average recall of 70.1% and an average F1-score of 71.3%, demonstrating high classification accuracy and being able to provide algorithmic support for arrhythmia classification in wearable devices.
OBJECTIVE In order to solve the difficult problem of one-stage repair of degloving injury of multiple fingers, the common pedicled ilio-inguinal-hypogastric subdermal vascular network skin flap was designed and the multi-lobes skin flap was performed subsequently. METHODS From 1993 to 1996, there were 5 cases with degloving injuries of multiple fingers were treated by this flap. There were 2 males and 3 females and the age ranged from 7 to 19 years old. RESULTS After operation, the pedicles of the flap was detached between 12 to 16 days and all of the flaps survived completely. Patients were followed up for 6-18 months. After repair, the contour and skin colour of the digits were excellent, and the motion of the interphalangeal joints and skin sensation were good. CONCLUSION The conclusion was as follows: The newly designed skin flap was characterized by the advantages of duration of treatment being short, excellent contour and more rapid recovery of function. It could be used for one-stage repair of degloving injury of multiple fingers.
