Automatic Population of Structured Reports from Narrative Pathology Reports

摘要

There are a number of advantages for the use of structured pathology reports: they can ensure the accuracy and completeness of pathology reporting; it is easier for the referring doctors to glean pertinent information from them.ududThe goal of this thesis is to extract pertinent information from free-text pathology reports and automatically populate structured reports for cancer diseases and identify the commonalities and differences in processing principles to obtain maximum accuracy.ududThree pathology corpora were annotated with entities and relationships between the entities in this study, namely the melanoma corpus, the colorectal cancer corpus and the lymphoma corpus. ududA supervised machine-learning based-approach, utilising conditional random fields learners, was developed to recognise medical entities from the corpora. By feature engineering, the best feature configurations were attained, which boosted the F-scores significantly from 4.2% to 6.8% on the training sets. ududWithout proper negation and uncertainty detection, the quality of the structured reports will be diminished. The negation and uncertainty detection modules were built to handle this problem. The modules obtained overall F-scores ranging from 76.6% to 91.0% on the test sets. ududA relation extraction system was presented to extract four relations from the lymphoma corpus. The system achieved very good performance on the training set, with 100% F-score obtained by the rule-based module and 97.2% F-score attained by the support vector machines classifier.ududRule-based approaches were used to generate the structured outputs and populate them to predefined templates. The rule-based system attained over 97% F-scores on the training sets. ududA pipeline system was implemented with an assembly of all the components described above. It achieved promising results in the end-to-end evaluations, with 86.5%, 84.2% and 78.9% F-scores on the melanoma, colorectal cancer and lymphoma test sets respectively.