Abstract | ||
---|---|---|
•Automatic extraction of morbid disease or conditions contained in death certificates is extremely useful for standardization, alleviating and smoothing human work. The positive impact of standardization is specially relevant for epidemiological studies, comparison across physicians, hospitals and countries and also for billing purposes.•General and multilingual approach to render diagnostic terms in death certificates into the standard framework provided by the ICD.•Automatic coding of diagnostic terms treated as an automatic translation task.•Study of the impact of different neural architectures on sequence-to-sequence ICD-10 coding.•Our results give a new state of the art on multilingual ICD-10 coding, outperforming several alternative approaches.•Informative ICD-10 coding, interpretable by clinicians. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1016/j.ijmedinf.2019.05.015 | International Journal of Medical Informatics |
Keywords | Field | DocType |
International Classification of Diseases,Electronic health records,Sequence-to-sequence mapping,Neural machine translation | Data mining,Terminology,Coding (social sciences),Natural language,Natural language processing,Artificial intelligence,Deep learning,Artificial neural network,Medicine,ICD-10,Encoding (memory) | Journal |
Volume | ISSN | Citations |
129 | 1386-5056 | 2 |
PageRank | References | Authors |
0.43 | 0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Aitziber Atutxa | 1 | 16 | 5.78 |
Arantza Díaz de Ilarraza | 2 | 60 | 24.18 |
Koldo Gojenola | 3 | 164 | 26.64 |
Maite Oronoz | 4 | 82 | 18.92 |
Olatz Perez-de-Viñaspre | 5 | 11 | 4.40 |