Temporal Segmentation for Capturing Snapshots of Patient Histories in Korean Clinical Narrative
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Objectives: Clinical discharge summaries provide valuable information about patients’ clinical history, which is helpful for the realization of intelligent healthcare applications. The documents tend to take the form of separate segments based on temporal or topical information. If a patient’s clinical history can be seen as a consecutive sequence of clinical events, then each temporal segment can be seen as a snapshot, providing a certain clinical context at a specific moment. This study aimed to demonstrate a temporal segmentation method of Korean clinical narratives for identifying textual snapshots of patient history as a proof-of-a-concept. Methods: Our method uses pattern-based segmentation to approximate human recognition of the temporal or topical shifts in clinical documents. We utilized rheumatic patients’ discharge summaries and transformed them into sequences of constituent chunks. We built 97 single pattern functions to denote whether a certain chunk has attributes that indicate that it can be a segment boundary. We manually defined the relationships between the pattern functions to resolve multiple pattern matchings and to make a final decision. Results: The algorithm segmented 30 discharge summaries and processed 1,849 decision points. Three human judges were asked whether they agreed with the algorithm’s prediction, and the agreement percentage on the judges’ majority opinion was 89.61%. Conclusions: Although this method is based on manually constructed rules, our findings demonstrate that the proposed algorithm can achieve fairly good segmentation results, and it may be the basis for methodological improvement in the future.
Abstract
Objectives: Clinical discharge summaries provide valuable information about patients’ clinical history, which is helpful for the realization of intelligent healthcare applications. The documents tend to take the form of separate segments based on temporal or topical information. If a patient’s clinical history can be seen as a consecutive sequence of clinical events, then each temporal segment can be seen as a snapshot, providing a certain clinical context at a specific moment. This study aimed to demonstrate a temporal segmentation method of Korean clinical narratives for identifying textual snapshots of patient history as a proof-of-a-concept. Methods: Our method uses pattern-based segmentation to approximate human recognition of the temporal or topical shifts in clinical documents. We utilized rheumatic patients’ discharge summaries and transformed them into sequences of constituent chunks. We built 97 single pattern functions to denote whether a certain chunk has attributes that indicate that it can be a segment boundary. We manually defined the relationships between the pattern functions to resolve multiple pattern matchings and to make a final decision. Results: The algorithm segmented 30 discharge summaries and processed 1,849 decision points. Three human judges were asked whether they agreed with the algorithm’s prediction, and the agreement percentage on the judges’ majority opinion was 89.61%. Conclusions: Although this method is based on manually constructed rules, our findings demonstrate that the proposed algorithm can achieve fairly good segmentation results, and it may be the basis for methodological improvement in the future.
참고문헌 (30)
- Zhou L, "System architecture for temporal information extraction, representation and reasoning in clinical narrative reports", AMIA Annu Symp Proc, 2005, pp.869~873 (2005)
- Combi C, "Temporal reasoning and temporal data maintenance in medicine: issues and challenges", Comput Biol Med, 27(5), pp.353~368 (1997)[DOI]
- Madkour M, "Temporal data representation, normalization, extraction, and reasoning: A review from clinical domain", Comput Methods Programs Biomed, 128, pp.52~68 (2016)[DOI]
- Verhagen M, "SemEval-2007 task 15: TempEval temporal relation identification", Proceedings of the 4th International Workshop on Semantic Evaluations, pp.75~80 (2007)
- Bramsen PJ, "Doing time: inducing temporal graphs", pp.2006
- Zhou L, "The evaluation of a temporal reasoning system in processing clinical discharge summaries", J Am Med Inform Assoc, 15(1), pp.99~106 (2008)[DOI]
- Pustejovsky J, "Increasing informativeness in temporal annotation", Proceedings of the 5th Linguistic Annotation Workshop, pp.152~160 (2011)
- Dligach D, "Neural temporal relation extraction", Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, 2, pp.746~751 (2017)
- Tao C, "Time-oriented question answering from clinical narratives using semantic-web techniques", International Semantic Web Conference, pp.241~256 (2010)
- Park H, "V-Model: a new perspective for EHRbased phenotyping", BMC Med Inform Decis Mak, 14, pp.90~ (2014)[DOI]
- Jung H, "Building timelines from narrative clinical records: initial results based-on deep natural language understanding", Proceedings of BioNLP 2011 Workshop, pp.146~154 (2011)
- Monroe M, "Temporal event sequence simplification", IEEE Trans Vis Comput Graph, 19(12), pp.2227~2236 (2013)[DOI]
- Angelova G, "Towards temporal segmentation of patient history in discharge letters", Proceedings of the Workshop on Biomedical Natural Language Processing, pp.49~54 (2011)
- Bramsen P, "Finding temporal order in discharge summaries", AMIA Annu Symp Proc, 2006, pp.81~85 (2006)
- Nakhimovsky A, "Discontinuities in narratives", Proceedings of the 12th Conference on Computational Linguistics, pp.465~470 (1988)
- Allen J, "Natural language understanding" (1995), Pearson
- Nakhimovsky A, "Aspect, aspectual class, and the temporal structure of narrative", Comput Linguist, 14(2), pp.29~43 (1988)
- Grosz BJ, "Attention, intentions, and the structure of discourse", Comput Linguist, 12(3), pp.175~204 (1986)
- Anderson A, "The accessibility of pronominal antecedents as a function of episode shifts in narrative text", Q J Exp Psychol A, 35(3), pp.427~440 (1983)[DOI]
- Maybury MT, "Using discourse focus, temporal focus, and spatial focus to generate multisentential text", Proceedings of the 5th International Workshop on Natural Language Generation, pp.70~78 (1990)
- Passonneau RJ, "Computational and conversational discourse", pp.161~194 (1996), Springer
- Marcu D, "The theory and practice of discourse parsing and summarization" (2000), MIT Press
- Beeferman D, "Text segmentation using exponential models", Proceedings of the 2nd Conference on Empirical Methods in Natural Language Processing, pp.37~46 (1997)
- Pevzner L, "A critique and improvement of an evaluation metric for text segmentation", Comput Linguist, 28(1), pp.19~36 (2002)[DOI]
- 김영호, "Recognizing Temporal Information in Korean Clinical Narratives through Text Normalization", Healthcare Informatics Research, 17(3), pp.150~155 (2011), 대한의료정보학회
- Chapman WW, "Overcoming barriers to NLP for clinical text: the role of shared tasks and the need for additional creative solutions", J Am Med Inform Assoc, 18(5), pp.540~543 (2011)[DOI]
- Seol JW, "Causality patterns and machine learning for the extraction of problem-action relations in discharge summaries", Int J Med Inform, 98, pp.1~12 (2017)[DOI]
- Uzuner O, "2010 i2b2", VA challenge on concepts, assertions, and relations in clinical text, 2011, pp.18(5), 552~556
- Sun W, "Evaluating temporal relations in clinical text: 2012 i2b2 Challenge", J Am Med Inform Assoc, 20(5), pp.806~813 (2013)[DOI]
- Bramsen P, "Inducing temporal graphs", Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing, pp.189~198 (2006)