HyperCore: Hyperbolic and Co-graph Representation for Automatic ICD Coding

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3105–3114 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 3105 HyperCore: Hyperbolic and Co-graph Representation for Automatic ICD Coding Pengfei Cao1,2, Yubo Chen1,2, Kang Liu1,2, Jun Zhao1,2, Shengping Liu3 and Weifeng Chong3 1 National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China 2 School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, 100049, China 3 Beijing Unisound Information Technology Co., Ltd, Beijing, 100028, China {pengfei.cao, yubo.chen, kliu, jzhao}@nlpr.ia.ac.cn, {liushengping, chongweifeng}@unisound.com Abstract The International Classification of Diseases (ICD) provides a standardized way for clas- sifying diseases, which endows each disease with a unique code. ICD coding aims to as- sign proper ICD codes to a medical record. Since manual coding is very laborious and prone to errors, many methods have been pro- posed for the automatic ICD coding task. How- ever, most of existing methods independently predict each code, ignoring two important characteristics: Code Hierarchy and Code Co-occurrence. In this paper, we propose a Hyperbolic and Co-graph Representation method (HyperCore) to address the above problem. Specifically, we propose a hyper- bolic representation method to leverage the code hierarchy. Moreover, we propose a graph convolutional network to utilize the code co-occurrence. Experimental results on two widely used datasets demonstrate that our proposed model outperforms previous state-of- the-art methods. 1 Introduction The International Classification of Diseases (ICD) is a healthcare classification system supported by the World Health Organization, which provides a unique code for each disease, symptom, sign and so on. ICD codes have been widely used for ana- lyzing clinical data and monitoring health issues (Choi et al., 2016; Avati et al., 2018). Due to the importance of ICD codes, ICD coding – which as- signs proper ICD codes to a medical record – has drawn much attention. The task of ICD coding is usually undertaken by professional coders ac- cording to doctors’ diagnosis descriptions in the form of free texts. However, manual coding is very expensive, time-consuming and error-prone. Automatic ICD Coding Model Mr.[**Known lastname 58216**] is an 87 year old male with Parkinsons Disease, difficulty breathing ,…,… 87 year old male presents with severe chest tightness, respiratory failure, and pneumatosis coli indicative of visceral necrosis. As the patient was not a surgical candidate, medical prognosis was poor …… Input: Clinical Text Output: Predicted ICD codes ICD-9 Codes Disease Name 518.81 Acute respiratory failure 401.9 Essential hypertension 276.2 Acidosis 038.9 Unspecified septicemia …… …… Figure 1: An example of automatic ICD coding task. The input and output of the automatic ICD coding model are clinical text and predicted ICD codes, respec- tively. For better understanding, we add the correspond- ing disease name for each code. The cost incurred by coding errors and the finan- cial investment spent on improving coding quality are estimated to be $25 billion per year in the US (Lang, 2007). Two main reasons can account for this. First, only the people who have medical …

Content truncated to 500 words