KFU. Card publication. An Encoder-Decoder Model for ICD-10 Coding of Death Certificates

KAZAN
FEDERAL UNIVERSITY

- RUS
- 中文
- ES

AN ENCODER-DECODER MODEL FOR ICD-10 CODING OF DEATH CERTIFICATES

Form of presentation

Conference proceedings in Russian journals and collections

Year of publication

2017

Язык

русский

Authors, employees of KFU

Tutubalina Elena Viktorovna, author

Bibliographic description in the original language

An Encoder-Decoder Model for ICD-10 Coding of Death Certificates // NIPS 2017 Machine Learning for Health Workshop. - 2017.

Annotation

NIPS 2017 Machine Learning for Health Workshop (NIPS ML4H 2017)

Keywords

deep learning, ICD coding, neural networks, encoder-decoder

The name of the journal

NIPS 2017 Machine Learning for Health Workshop (NIPS ML4H 2017)

Please use this ID to quote from or refer to the card

https://repository.kpfu.ru/eng/?p_id=168960&p_lang=2

Resource files

File name	Size (MB)	Format
F_encoder_decoder_model_3.pdf	0,12	pdf	show / download

Full metadata record

Field DC	Value	Language
dc.contributor.author	Tutubalina Elena Viktorovna	ru_RU
dc.date.accessioned	2017-01-01T00:00:00Z	ru_RU
dc.date.available	2017-01-01T00:00:00Z	ru_RU
dc.date.issued	2017	ru_RU
dc.identifier.citation	An Encoder-Decoder Model for ICD-10 Coding of Death Certificates // NIPS 2017 Machine Learning for Health Workshop. - 2017.	ru_RU
dc.identifier.uri	https://repository.kpfu.ru/eng/?p_id=168960&p_lang=2	ru_RU
dc.description.abstract	NIPS 2017 Machine Learning for Health Workshop (NIPS ML4H 2017)	ru_RU
dc.description.abstract	Information extraction from textual documents such as hospital records and health-related user discussions has become a topic of intense interest. The task of medical concept coding is to map a variable length text to medical concepts and corresponding classification codes in some external system or ontology. In this work, we utilize recurrent neural networks to automatically assign ICD-10 codes to fragments of death certificates written in English. We develop end-to-end neural architectures directly tailored to the task, including basic encoder-decoder architecture for statistical translation. In order to incorporate prior knowledge, we concatenate cosine similarities vector among the text and dictionary entry to the encoded state. Being applied to a standard benchmark from CLEF eHealth 2017 challenge, our model achieved F-measure of 85.01\% on a full test set with significant improvement as compared to the average score of 62.2\% for all official participants' approaches.	ru_RU
dc.language.iso	ru	ru_RU
dc.subject	deep learning	ru_RU
dc.subject	ICD coding	ru_RU
dc.subject	neural networks	ru_RU
dc.subject	encoder-decoder	ru_RU
dc.title	An Encoder-Decoder Model for ICD-10 Coding of Death Certificates	ru_RU
dc.type	Conference proceedings in Russian journals and collections	ru_RU