Retinopathy of prematurity (ROP) is a disease that can cause blindness in premature infants. It is characterized by immature vascular growth of the retinal blood vessels. However, early detection and treatment of ROP can significantly improve the visual acuity of high-risk patients. Thus, early diagnosis of ROP is crucial in preventing visual impairment. However, several patients refrain from treatment owing to the lack of medical expertise in diagnosing the disease; this is especially problematic considering that the number of ROP cases is on the rise. To this end, we applied transfer learning to five deep neural network architectures for identifying ROP in preterm infants. Our results showed that the VGG19 model outperformed the other models in determining whether a preterm infant has ROP, with 96% accuracy, 96.6% sensitivity, and 95.2% specificity. We also classified the severity of the disease; the VGG19 model showed 98.82% accuracy in predicting the severity of the disease with a sensitivity and specificity of 100% and 98.41%, respectively. We performed 5-fold cross-validation on the datasets to validate the reliability of the VGG19 model and found that the VGG19 model exhibited high accuracy in predicting ROP. These findings could help promote the development of computer-aided diagnosis.

Sound classification has been widely used in many fields. Unlike traditional signal-processing methods, using deep learning technology for sound classification is one of the most feasible and effective methods. However, limited by the quality of the training dataset, such as cost and resource constraints, data imbalance, and data annotation issues, the classification performance is affected. Therefore, we propose a sound classification mechanism based on convolutional neural networks and use the sound feature extraction method of Mel-Frequency Cepstral Coefficients (MFCCs) to convert sound signals into spectrograms. Spectrograms are suitable as input for CNN models. To provide the function of data augmentation, we can increase the number of spectrograms by setting the number of triangular bandpass filters. The experimental results show that there are 50 semantic categories in the ESC-50 dataset, the types are complex, and the amount of data is insufficient, resulting in a classification accuracy of only 63%. When using the proposed data augmentation method (K = 5), the accuracy is effectively increased to 97%. Furthermore, in the UrbanSound8K dataset, the amount of data is sufficient, so the classification accuracy can reach 90%, and the classification accuracy can be slightly increased to 92% via data augmentation. However, when only 50% of the training dataset is used, along with data augmentation, the establishment of the training model can be accelerated, and the classification accuracy can reach 91%.

The development of IoT technology has made various IoT applications and services widely used. Because IoT devices have weak information security protection capabilities, they are easy targets for cyber attacks. Therefore, this study proposes MLP-based IoT attack classification with data augmentation for GANs. In situations where the overall classification performance is satisfactory but the performance of a specific class is poor, GANs are employed as a data augmentation mechanism for that class to enhance its classification performance. The experimental results indicate that regardless of whether the training dataset is BoT-IoT or TON-IOT, the proposed method significantly improves the classification performance of classes with insufficient training data when using the data augmentation mechanism with GANs. Furthermore, the classification accuracy, precision, recall, and F1-score performance all exceed 90%.

In recent years, within large cities with a high population density, traffic congestion has become more and more serious, resulting in increased emissions of vehicles and reducing the efficiency of urban operations. Many factors have caused traffic congestion, such as insufficient road capacity, high vehicle density, poor urban traffic planning and inconsistent traffic light cycle configuration. Among these factors, the problems of traffic light cycle configuration are the focal points of this paper. If traffic lights can adjust the cycle dynamically with traffic data, it will reduce degrees of traffic congestion significantly. Therefore, a modified mechanism based on Q-Learning to optimize traffic light cycle configuration is proposed to obtain lower average vehicle delay time, while keeping significantly fewer processing steps. The experimental results will show that the number of processing steps of this proposed mechanism is 11.76 times fewer than that of the exhaustive search scheme, and also that the average vehicle delay is only slightly lower than that of the exhaustive search scheme by 5.4%. Therefore the proposed modified Q-learning mechanism will be capable of reducing the degrees of traffic congestions effectively by minimizing processing steps.