A novel bidirectional LSTM network based on scale factor for atrial fibrillation signals classification

Compressed Deep Learning Models for Wearable Atrial Fibrillation Detection through Attention

Deep learning (DL) models have shown promise for the accurate detection of atrial fibrillation (AF) from electrocardiogram/photoplethysmography (ECG/PPG) data, yet deploying these on resource-constrained wearable devices remains challenging. This study proposes integrating a customized channel attention mechanism to compress DL neural networks for AF detection, allowing the model to focus only on the most salient time-series features. The results demonstrate that applying compression through channel attention significantly reduces the total number of model parameters and file size while minimizing loss in detection accuracy. Notably, after compression, performance increases for certain model variants in key AF databases (ADB and C2017DB). Moreover, analyzing the learned channel attention distributions after training enhances the explainability of the AF detection models by highlighting the salient temporal ECG/PPG features most important for its diagnosis. Overall, this research establishes that integrating attention mechanisms is an effective strategy for compressing large DL models, making them deployable on low-power wearable devices. We show that this approach yields compressed, accurate, and explainable AF detectors ideal for wearables. Incorporating channel attention enables simpler yet more accurate algorithms that have the potential to provide clinicians with valuable insights into the salient temporal biomarkers of AF. Our findings highlight that the use of attention is an important direction for the future development of efficient, high-performing, and interpretable AF screening tools for wearable technology.

Research on Multimodal Dance Movement Recognition Based on Artificial Intelligence Image Technology

At present, most robot dances are precompiled. Changing music requires manual adjustment of relevant parameters and metamovements, which greatly reduces the fun and intelligence. In view of the above problems, this paper designed CNN system, studied the multimodal dance movement recognition algorithm of artificial intelligence image technology, and completed the construction of a multimodal dance movement calculation system example. The results show that the CNN algorithm and the Winograd algorithm-based coprocessor-optimized CNN network in multimodal dance movement recognition with image technology reduce from a maximum of 132s to 26s in the runtime criterion, with a maximum reduction of 80%; from a maximum of 73.5% to 16.2% in the memory access criterion, with a maximum reduction of 57.3%; and from a maximum of 93.6% to 25.2% in the power consumption ratio criterion, with a maximum reduction of 68.4%. In the power consumption ratio criterion, the maximum reduction from 93.6% to 25.2% is 68.4%. The maximum accuracy of the proposed optimization method is 95.1%. The solution is proposed to address the problem of insufficient performance of traditional dance movement recognition, which will contribute to the development of artificial intelligence and dance industry.