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.
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.
Manual segmentation of coronary arteries in computed tomography angiography (CTA) images is inefficient, and existing deep learning segmentation models often exhibit low accuracy on coronary artery images. Inspired by the Transformer architecture, this paper proposes a novel segmentation model, the double parallel encoder u-net with transformers (DUNETR). This network employed a dual-encoder design integrating Transformers and convolutional neural networks (CNNs). The Transformer encoder transformed three-dimensional (3D) coronary artery data into a one-dimensional (1D) sequential problem, effectively capturing global multi-scale feature information. Meanwhile, the CNN encoder extracted local features of the 3D coronary arteries. The complementary features extracted by the two encoders were fused through the noise reduction feature fusion (NRFF) module and passed to the decoder. Experimental results on a public dataset demonstrated that the proposed DUNETR model achieved a Dice similarity coefficient of 81.19% and a recall rate of 80.18%, representing improvements of 0.49% and 0.46%, respectively, over the next best model in comparative experiments. These results surpassed those of other conventional deep learning methods. The integration of Transformers and CNNs as dual encoders enables the extraction of rich feature information, significantly enhancing the effectiveness of 3D coronary artery segmentation. Additionally, this model provides a novel approach for segmenting other vascular structures.
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.
Electrical impedance tomography (EIT) is a non-radiation, non-invasive visual diagnostic technique. In order to improve the imaging resolution and the removing artifacts capability of the reconstruction algorithms for electrical impedance imaging in human-chest models, the HMANN algorithm was proposed using the Hadamard product to optimize multilayer artificial neural networks (MANN). The reconstructed images of the HMANN algorithm were compared with those of the generalized vector sampled pattern matching (GVSPM) algorithm, truncated singular value decomposition (TSVD) algorithm, backpropagation (BP) neural network algorithm, and traditional MANN algorithm. The simulation results showed that the correlation coefficient of the reconstructed images obtained by the HMANN algorithm was increased by 17.30% in the circular cross-section models compared with the MANN algorithm. It was increased by 13.98% in the lung cross-section models. In the lung cross-section models, some of the correlation coefficients obtained by the HMANN algorithm would decrease. Nevertheless, the HMANN algorithm retained the image information of the MANN algorithm in all models, and the HMANN algorithm had fewer artifacts in the reconstructed images. The distinguishability between the objects and the background was better compared with the traditional MANN algorithm. The algorithm could improve the correlation coefficient of the reconstructed images, and effectively remove the artifacts, which provides a new direction to effectively improve the quality of the reconstructed images for EIT.
Cardiotocography (CTG) is a non-invasive and important tool for diagnosing fetal distress during pregnancy. To meet the needs of intelligent fetal heart monitoring based on deep learning, this paper proposes a TWD-MOAL deep active learning algorithm based on the three-way decision (TWD) theory and multi-objective optimization Active Learning (MOAL). During the training process of a convolutional neural network (CNN) classification model, the algorithm incorporates the TWD theory to select high-confidence samples as pseudo-labeled samples in a fine-grained batch processing mode, meanwhile low-confidence samples annotated by obstetrics experts were also considered. The TWD-MOAL algorithm proposed in this paper was validated on a dataset of 16 355 prenatal CTG records collected by our group. Experimental results showed that the algorithm proposed in this paper achieved an accuracy of 80.63% using only 40% of the labeled samples, and in terms of various indicators, it performed better than the existing active learning algorithms under other frameworks. The study has shown that the intelligent fetal heart monitoring model based on TWD-MOAL proposed in this paper is reasonable and feasible. The algorithm significantly reduces the time and cost of labeling by obstetric experts and effectively solves the problem of data imbalance in CTG signal data in clinic, which is of great significance for assisting obstetrician in interpretations CTG signals and realizing intelligence fetal monitoring.
Focus on the inconsistency of the shape, location and size of brain glioma, a dual-channel 3-dimensional (3D) densely connected network is proposed to automatically segment brain glioma tumor on magnetic resonance images. Our method is based on a 3D convolutional neural network frame, and two convolution kernel sizes are adopted in each channel to extract multi-scale features in different scales of receptive fields. Then we construct two densely connected blocks in each pathway for feature learning and transmission. Finally, the concatenation of two pathway features was sent to classification layer to classify central region voxels to segment brain tumor automatically. We train and test our model on open brain tumor segmentation challenge dataset, and we also compared our results with other models. Experimental results show that our algorithm can segment different tumor lesions more accurately. It has important application value in the clinical diagnosis and treatment of brain tumor diseases.
Fetal electrocardiogram (ECG) signals provide important clinical information for early diagnosis and intervention of fetal abnormalities. In this paper, we propose a new method for fetal ECG signal extraction and analysis. Firstly, an improved fast independent component analysis method and singular value decomposition algorithm are combined to extract high-quality fetal ECG signals and solve the waveform missing problem. Secondly, a novel convolutional neural network model is applied to identify the QRS complex waves of fetal ECG signals and effectively solve the waveform overlap problem. Finally, high quality extraction of fetal ECG signals and intelligent recognition of fetal QRS complex waves are achieved. The method proposed in this paper was validated with the data from the PhysioNet computing in cardiology challenge 2013 database of the Complex Physiological Signals Research Resource Network. The results show that the average sensitivity and positive prediction values of the extraction algorithm are 98.21% and 99.52%, respectively, and the average sensitivity and positive prediction values of the QRS complex waves recognition algorithm are 94.14% and 95.80%, respectively, which are better than those of other research results. In conclusion, the algorithm and model proposed in this paper have some practical significance and may provide a theoretical basis for clinical medical decision making in the future.
With the rapid improvement of the perception and computing capacity of mobile devices such as smart phones, human activity recognition using mobile devices as the carrier has been a new research hot-spot. The inertial information collected by the acceleration sensor in the smart mobile device is used for human activity recognition. Compared with the common computer vision recognition, it has the following advantages: convenience, low cost, and better reflection of the essence of human motion. Based on the WISDM data set collected by smart phones, the inertial navigation information and the deep learning algorithm-convolutional neural network (CNN) were adopted to build a human activity recognition model in this paper. The K nearest neighbor algorithm (KNN) and the random forest algorithm were compared with the CNN network in the recognition accuracy to evaluate the performance of the CNN network. The classification accuracy of CNN model reached 92.73%, which was much higher than KNN and random forest. Experimental results show that the CNN algorithm model can achieve more accurate human activity recognition and has broad application prospects in predicting and promoting human health.
The measurement of network is one of the important researches in resolving neuronal population information processing mechanism using complex network theory. For the quantitative measurement problem of functional neural network, the relation between the measure indexes, i.e. the clustering coefficient, the global efficiency, the characteristic path length and the transitivity, and the network topology was analyzed. Then, the spike-based functional neural network was established and the simulation results showed that the measured network could represent the original neural connections among neurons. On the basis of the former work, the coding of functional neural network in nidopallium caudolaterale (NCL) about pigeon's motion behaviors was studied. We found that the NCL functional neural network effectively encoded the motion behaviors of the pigeon, and there were significant differences in four indexes among the left-turning, the forward and the right-turning. Overall, the establishment method of spike-based functional neural network is available and it is an effective tool to parse the brain information processing mechanism.
