Search

The task of assigning diagnostic ICD codes to patient hospital admissions is typically performed by expert human coders. Efforts towards automated ICD coding are dominated by supervised deep learning models. However, difficulties in learning to predict the large number of rare codes remain a barrier to adoption in clinical practice. In this work, we leverage off-the-shelf pre-trained generative large language models (LLMs) to develop a practical solution that is suitable for zero-shot and few-shot code assignment, with no need for further task-specific training. Unsupervised pre-training alone does not guarantee precise knowledge of the ICD ontology and specialist clinical coding task, therefore we frame the task as information extraction, providing a description of each coded concept and asking the model to retrieve related mentions. For efficiency, rather than iterating over all codes, we leverage the hierarchical nature of the ICD ontology to sparsely search for relevant codes., Comment: Accepted to the NeurIPS 2023 workshop Deep Generative Models For Health (DGM4H). 9 pages, 3 figures

Reviews use of mixed-code teaching in Hong Kong schools and outlines educators' efforts to persuade the government to accept mixed-code as not only inevitable, but preferable in most Hong Kong schools. Notes that English must be preserved for business purposes, but the Chinese medium of instruction is receiving greater emphasis as Hong Kong nears the date for its takeover by China. (25 references) (Author/CK)

Considers the role of native-speaker English language teachers in Hong Kong. First, the article examines the concept of native speaker and offers five criteria that determine whether a person can be classed as a native speaker or not. Second, it focuses on the use of native-speaker English teachers in Hong Kong, with particular reference to a project known as the Expatriates English Teachers Scheme. (JL)

Lists and discusses factors affecting listening comprehension. List was compiled from a literature survey, consulting a cross-section of teachers working at secondary and tertiary levels, and questioning students who had just finished their secondary education. (EKN)

The results of a survey of 180 part-time students' opinions on the standard of English in Hong Kong, on governmental efforts to improve the level of English in Hong Kong, and on how returning Hong Kong to China in 1997 will effect the standard of English are reported. The students were enrolled in a 6-year course in Chinese/English/Translation at the Chinese University of Hong Kong. The questionnaire used in the survey included opportunities for both open and closed responses so that both broad and specific opinions would be reported. A clear majority, 55%, felt that the standard of English in Hong Kong is poor, while only 4% considered the level of English in Hong Kong to be good. Fifty-six percent of the respondents felt that the standard of English has declined. The most frequently offered evidence for this answer was the poorer command of English among younger colleagues of the respondents in work situations. The vast majority of the respondents, especially teachers, were unhappy about the English teaching situation (unpopular job, unmotivated students) in the schools. Over 50% of the respondents felt that each of the governmental proposals for improving the status of English in Hong Kong (the Expatriate Teachers Scheme, sending local teachers abroad, or a policy of "positive discrimination" favoring schools which move from English to Chinese as the medium of instruction) should be tried. Finally, a fairly general consensus was reached on the point that, although English will be needed for such activities as international business and tourism after 1997, the use and status of English in Hong Kong will decline. (DJD)