Using Patent Drawings of Invention Publications to Differentiate Stock Return Rate - an Empirical Study on China Stock Market

Patent is an important outcome of technological innovation, which drove economic growth. Though patent claim always caught attention when considering patentability, it had to be supported by the drawings according to the patent examination criteria. However, patent drawing was seldom discussed. More than 60% of China listed companies of RMB common stocks (A-shares) from 2017Q1 to 2021Q4 were selected as effective samples based on the company integrated database. The effect of China invention publication's patent drawing count for differentiating China A-share's stock return rate was thoroughly discussed via analysis of variation (ANOVA). The average drawing count and the total drawing count of invention publications significantly increased over previous five years even under the impact of COVID-19 pandemic. Moreover, the total drawing count was found to be an appropriate patent indicator for differentiating A-share's stock return rate whereas the average drawing count was not because the average drawing count did not show significant connection with the stock return rate. The Ashares in the group of the highest total drawing counts showed significantly higher stock return rate mean while the A-shares in the groups of lower total drawing counts showed significantly lower stock return rate means in most quarters from 2017Q1 to 2021Q4. The finding also proved that the patent quantity still mattered in China stock market.

Multi-feature Multi-Scale CNN-Derived COVID-19 Classification from Lung Ultrasound Data.

The global pandemic of the novel coronavirus disease 2019 (COVID-19) has put tremendous pressure on the medical system. Imaging plays a complementary role in the management of patients with COVID-19. Computed tomography (CT) and chest X-ray (CXR) are the two dominant screening tools. However, difficulty in eliminating the risk of disease transmission, radiation exposure and not being cost-effective are some of the challenges for CT and CXR imaging. This fact induces the implementation of lung ultrasound (LUS) for evaluating COVID-19 due to its practical advantages of noninvasiveness, repeatability, and sensitive bedside property. In this paper, we utilize a deep learning model to perform the classification of COVID-19 from LUS data, which could produce objective diagnostic information for clinicians. Specifically, all LUS images are processed to obtain their corresponding local phase filtered images and radial symmetry transformed images before fed into the multi-scale residual convolutional neural network (CNN). Secondly, image combination as the input of the network is used to explore rich and reliable features. Feature fusion strategy at different levels is adopted to investigate the relationship between the depth of feature aggregation and the classification accuracy. Our proposed method is evaluated on the point-of-care US (POCUS) dataset together with the Italian COVID-19 Lung US database (ICLUS-DB) and shows promising performance for COVID-19 prediction.