Ontological features of Electronic Health Records reveal distinct association patterns in liver cancer

Abstract

Electronic Health Record (EHR) system is not only aimed to provide a digital and structural form of patient records but also support the clinical decision, patient care and patient advice. The EHR database is still an under-explored big data resource that has hosted a large number of cases with complete recovery, good prognosis, reliable diagnostic tests and effective treatments. A set of 112 abdominal computed tomography imaging examination reports, consisting of 59 cases of hepatocellular carcinoma (HCC) or liver metastases (so called HCC group for simplicity) and 53 cases with no abnormality detected (NAD group), was collected from four hospitals in Hong Kong. We extracted terms related to liver cancer from the reports and mapped them to ontological features using Systematized Nomenclature of Medicine (SNOMED) Clinical Terms (CT). Each feature value was further weighted using a systematic PubMed search method. Association levels between every two features in HCC and NAD groups were quantified using Pearson's correlation coefficient. The distribution of association levels in HCC group was compared with that in NAD group. HCC group reveals a distinct association pattern that signifies liver cancer and provides clinical decision support for suspected cases.