COVID-19 impact on the Chinese top academic libraries: Libraries' response to space, collection and services.

In 2020, the COVID-19 pandemic was a major public health emergency on a global scale. The literature regarding the pandemic and its impact on academic libraries is still rising. This article examines the two-year process of developing a flexible service scenario and the broader picture by analyzing data on Chinese top university libraries' programmes and outreach initiatives prior to, during, and the normal COVID-19 pandemic (Sept. 2019-Sept. 2021). COVID-19 is found to have a significant impact on the physical space, collection development, and service of the library, demonstrating the characteristics of space access restricted by security measures, collection digitization, and online service. This research also examines the previous year's initiatives and programmes and discusses the next phase of "new normal" procedures. Hopefully, this study will give insight on how Chinese libraries responded to the recent pandemic, informing libraries' outreach and efforts to be better prepared to take imperative, swift, and decisive action in the post-COVID-19 era and beyond.

Layout-aware information extraction from semi-structured medical images.

Textual information embedded in the medical image contains rich structured information about the medical condition of a patient. This paper aims at extracting structured textual information from semi-structured medical images. Given the recognized text spans of an image preprocessed by optical character recognition (OCR), due to the spatial discontinuity of texts spans as well as potential errors brought by OCR, the structured information extraction becomes more challenging. In this paper, we propose a domain-specific language, called ODL, which allows users to describe the value and layout of text data contained in the images. Based on the value and spatial constraints described in ODL, the ODL parser associates values found in the image with the data structure in the ODL description, while conforming to the aforementioned constraints. We conduct experiments on a dataset consisting of real medical images, our ODL parser consistently outperforms existing approaches in terms of extraction accuracy, which shows the better tolerance of incorrectly recognized texts, and positional variances between images. This accuracy can be further improved by learning from a few manual corrections.