Automatic recognition of disorders, findings, pharmaceuticals and body structures from clinical text: An annotation and machine learning study

Abstract

Automatic recognition of clinical entities in the narrative text of health records is useful for constructing applications for documentation of patient care, as well as for secondary usage in the form of medical knowledge extraction. There are a number of named entity recognition studies on English clinical text, but less work has been carried out on clinical text in other languages. This study was performed on Swedish health records, and focused on four entities that are highly relevant for constructing a patient overview and for medical hypothesis generation, namely the entities: Disorder, Finding, Pharmaceutical Drug and Body Structure. The study had two aims: to explore how well named entity recognition methods previously applied to English clinical text perform on similar texts written in Swedish; and to evaluate whether it is meaningful to divide the more general category Medical Problem, which has been used in a number of previous studies, into the two more granular entities, Disorder and Finding. Clinical notes from a Swedish internal medicine emergency unit were annotated for the four selected entity categories, and the inter-annotator agreement between two pairs of annotators was measured, resulting in an average F-score of 0.79 for Disorder, 0.66 for Finding, 0.90 for Pharmaceutical Drug and 0.80 for Body Structure. A subset of the developed corpus was thereafter used for finding suitable features for training a conditional random fields model. Finally, a new model was trained on this subset, using the best features and settings, and its ability to generalise to held-out data was evaluated. This final model obtained an F-score of 0.81 for Disorder, 0.69 for Finding, 0.88 for Pharmaceutical Drug, 0.85 for Body Structure and 0.78 for the combined category Disorder+Finding. The obtained results, which are in line with or slightly lower than those for similar studies on English clinical text, many of them conducted using a larger training data set, show that the approaches used for English are also suitable for Swedish clinical text. However, a small proportion of the errors made by the model are less likely to occur in English text, showing that results might be improved by further tailoring the system to clinical Swedish. The entity recognition results for the individual entities Disorder and Finding show that it is meaningful to separate the general category Medical Problem into these two more granular entity types, e.g. for knowledge mining of co-morbidity relations and disorder-finding relations.

Enriching the international clinical nomenclature with Chinese daily used synonyms and concept recognition in physician notes

“…Each CTCF feature could be marked ‘positive’, ‘negative’ or ‘not-mentioned’ by scanning HPI, as a Bag-of-Concepts model instead of Bag-of-Words model [32]. Besides, the CTCFs from inpatients could be used for knowledge mining of co-morbidity relations and disorder-finding relations [33]. …”

An advanced review on text mining in medicine

“…Named entity recognition Benton et al (2011), Carrero, Cortizo, andGomez (2008), Ferrandez, South, Shen, and Meystre (2012), Kipper-Schuler, Kaggal, Masanz, Ogren, and Savova (2008), Lin, Chen, and Brown (2013), Rink, Harabagiu, and Roberts (2011), Roberts and Harabagiu (2011), Roberts, Rink, and Harabagiu (2013), Skeppstedt, Kvist, Nilsson, and Dalianis (2014), Uzuner, Mailoa, Ryan, and Sibanda (2010), Wang and Patrick (2009), Xia et al (2013), Zhu, Cherry, Kiritchenko, Martin, and De Bruijn (2013) Hypothesis generation and knowledge discovery Baron et al (2013), Byrd, Steinhubl, Sun, Ebadollahi, and Stewart (2014), Cole et al (2013) (2013) Text classification Asghar et al (2013), Castro et al (2015), Frunza, Inkpen, Matwin, Klement, andO'blenis (2011), Goldstein, Arzumtsyan, andUzuner (2007), Harpaz et al (2014), Jonnagaddala, Dai, Ray, and Liaw (2015), Metais, Nakache, and Timsit (2006), Pakhomov et al (2007), Vijayakrishnan et al (2014), Yetisgen-Yildiz and Pratt (2005), and Zuccon et al (2013) used including MetaMap (Aronson & Lang, 2010), Wikipedia, and WordNet. The resulting model was evaluated according to the 2010 i2b2/VA challenge data (Workshop on NLP Challenges for Clinical Records), obtaining a F-measure value of 0.796 for the concept extraction task (the best i2b2 submission value was 0.852, and the median value for all the submissions was 0.778).…”

Application of text mining in the biomedical domain

“…Typically, TM system that are based on algorithms such as hidden Markov models (HMM) [57,58], maximum entropy Markov models (MEMM) [59], conditional random fields (CRF) [60,61], and support vector machines (SVM) [62,63], need to be trained on a carefully constructed annotated training data set that is representative for the real life data set before the actual NER task. Machine learning based TM systems are used for instance to identify chemical entities in text [64], or are used in combination with rule-based and lexical methods to identify organism names in text [65] or used for extraction of cancer staging information from health records to improve clinical decision making [66].…”