Vasectomy reversal in China during the recent decade: insights from a multicenter retrospective investigation.

To date, there is little information about the demography of vasectomy reversal (VR) patients or the factors currently influencing VR effectiveness in China, especially after the universal two-child policy was released in 2015. In this research, demographic data and perioperative medical records of VR patients were extracted from seven major hospitals in different provinces or municipalities of China. Meanwhile, a telephone survey of the patients was conducted to collect follow-up information. Eventually, 448 VR cases from the past 13 years were included. The results were analyzed by stratified comparison to investigate factors that can influence postoperative vas deferens patency and pregnancy rate. Appropriately statistical methods were used, and all of the protocols were approved by the Ethics Committees of the institutes in this research. The results showed that the annual operation volume of VR quadrupled after the two-child policy was implemented. Nonmicrosurgery and a long duration of vasectomy were significantly associated with a lower patency rate. A follow-up survey showed that the general postoperative pregnancy rate was 27.2%. For female partners over the age of 35 years, the postoperative pregnancy rate showed a more severe decline, but only 35.5% of them had been given a fertility examination before their husbands' VR surgery. Our work revealed that more patients in China have been demanding VR in recent years. High-quality microsurgery and a short duration of vasectomy are crucial for restoring patency by VR. Clinical andrologists should perform a preoperative fertility evaluation of the patients' female partners.

CAPTCHA recognition based on deep convolutional neural network.

Aiming at the problems of low efficiency and poor accuracy of traditional CAPTCHA recognition methods, we have proposed a more efficient way based on deep convolutional neural network (CNN). The Dense Convolutional Network (DenseNet) has shown excellent classification performance which adopts cross-layer connection. Not only it effectively alleviates the vanishing-gradient problem, but also dramatically reduce the number of parameters. However, it also has caused great memory consumption. So we improve and construct a new DenseNet for CAPTCHA recognition (DFCR). Firstly, we reduce the number of convolutional blocks and build corresponding classifiers for different types of CAPTCHA images. Secondly, we input the CAPTCHA images of TFrecords format into the DFCR for model training. Finally, we test the Chinese or English CAPTCHAs experimentally with different numbers of characters. Experiments show that the new network not only keeps the primary performance advantages of the DenseNets but also effectively reduces the memory consumption. Furthermore, the recognition accuracy of CAPTCHA with the background noise and character adhesion is above 99.9%.