(1) Background and objective: Cardiovascular disease is one of the most common causes of death in today’s world. ECG is crucial in the early detection and prevention of cardiovascular disease. In this study, an improved deep learning method is proposed to diagnose abnormal and normal ECG accurately. (2) Methods: This paper proposes a CNN-FWS that combines three convolutional neural networks (CNN) and recursive feature elimination based on feature weights (FW-RFE), which diagnoses abnormal and normal ECG. F1 score and Recall are used to evaluate the performance. (3) Results: A total of 17,259 records were used in this study, which validated the diagnostic performance of CNN-FWS for normal and abnormal ECG signals in 12 leads. The experimental results show that the F1 score of CNN-FWS is 0.902, and the Recall of CNN-FWS is 0.889. (4) Conclusion: CNN-FWS absorbs the advantages of convolutional neural networks (CNN) to obtain three parts of different spatial information and enrich the learned features. CNN-FWS can select the most relevant features while eliminating unrelated and redundant features by FW-RFE, making the residual features more representative and effective. The method is an end-to-end modeling approach that enables an adaptive feature selection process without human intervention.
Most existing research achievements of digital watermarking techniques are in transform domain. In comparison with spatial domain, its advantages are larger data volume, higher security and stronger robustness. But its limitations are also obvious: complex computing requirement, weak in resisting attack and anti-extraction. In this paper, a novel blind digital watermarking algorithm is proposed, which performs digital watermark embedding process in Compressive Sensing (CS) domain based on the characteristics of CS and Human Visual System (HVS). The sub-blocks with larger capacity are selected to embed the scrambled digital watermark, considering the non-uniformity of blocks. Besides that, suitable quantization steps are chose adaptively by using quantization method. Experimental results show that the algorithm obtains robust and invisible embedded watermark with larger capacity of data. At the same time, the ability of defending against attack or extraction of embedded watermark is greatly improved. Most important feature in our algorithm is that the watermark can be extracted without any reference to the original image. As a result, the cost of storing carrier data can be saved remarkably.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.