Fatigue is a common state of mankind characterized by a reduction in the level of consciousness and alertness. Therefore, the recognition of fatigue and sleepiness has become indispensable in many alertness-dependent situations, such as when driving vehicles on public roads, performing demanding tasks in the workplace, or monitoring intensive care unit patients. This study proposes a method based on novel multi-feature fusion to detect fatigue and sleepiness by using traditional image processing and heart rate variability (HRV). The proposed method performs initial feature extraction using InceptionV3 (a convolutional neural network (CNN)), following which the second decision is made by a long short-term memory network (LSTM) using the features collected by InceptionV3 to process the sequence of video data for recognition. The LSTM provides coherent and precise sequence recognition that avoids static distortions. Then, the final decision is made by the blood volume pulse vector (PBV) method after the features are fused. Because fatigue recognition is usually employed to monitor driver fatigue, we verified the feasibility of our method by testing its ability to successfully recognize driver fatigue. Following the experiments, we compared the different steps in the proposed method with those in existing methods. We selected four other methods to perform the comparison tests and used the same videos for training networks. In comparison with state-of-the-art methods, our method in its entirety achieved an average increase of 5% in terms of both accuracy and stability.INDEX TERMS fatigue driving, LSTM network, convolutional neural network, blood volume pulse vector, blood volume pulse, heart rate variability.

Coronary artery calcification affects the arteries that supply the heart with blood, and percutaneous coronary intervention (PCI) is a direct and effective surgery to alleviate this symptom. In this paper, we propose a framework to judge if a patient requires surgery, based on cardiac computerized tomography scans. We adopt generative adversarial network to segment the calcified areas from slices. This architecture provides an environment for the generator to perform joint learning from ground truth images and the high-resolution discriminator. We use images reconstructed using two types of filters to test our method. An F1 score of 96.1% and 85.0% was achieved for the soft and sharp filters. In addition, we explored different recurrent neural networks for making the final decision. Including long short-term memory, which was ultimately used to deal with the calcium score normalized by the age and score threshold. Using the soft reconstruction image as the input, the whole framework achieved an accuracy of 76.6%. These results certify that our method can precisely locate lesion in artery, and make a reasonable risk assessment for PCI.INDEX TERMS Generative adversarial network, low-dose cardiac CT, recurrent neural network, percutaneous coronary intervention, coronary calcium scoring.

No abstract