• TUCS Publication Series
• How to Publish in TUCS Series
• Publication Search
• Publication Graph
• Publication Input
• Publication Input Guide
• JuFo Browser

Machine Learning to Automate the Assignment of Diagnosis Codes to Free-text Radiology Reports: a Method Description

Hanna Suominen, Filip Ginter, Sampo Pyysalo, Antti Airola, Tapio Pahikkala, Sanna Salanterä, Tapio Salakoski, Machine Learning to Automate the Assignment of Diagnosis Codes to Free-text Radiology Reports: a Method Description. In: Dale Schuurmans Csaba Szepesvari Milos Hauskrecht (Ed.), Proceedings of the ICML/UAI/COLT 2008 Workshop on Machine Learning for Health-Care Applications; 2008 July 9: Helsinki, Finland, 2008.

Abstract:

We introduce a multi-label classification system for the automated assignment of diagnostic codes to radiology reports. The system is a cascade of text enrichment, feature selection and two classifiers. It was evaluated in the Computational Medicine Center’s 2007 Medical Natural Language Processing Challenge and achieved a 87.7% micro-averaged F1-score and third place out of 44 submissions in the task, where 45 different ICD-9-CM codes were present in 94 combinations.
Especially the text enrichment and feature selection components are shown to contribute to our success. Our study provides insight
into the development of applications for real-life usage, which are currently rare.

Files:

Abstract in PDF-format

BibTeX entry:

@INPROCEEDINGS{inpSuGiPyAiPaSaSa08a,
  title = {Machine Learning to Automate the Assignment of Diagnosis Codes to Free-text Radiology Reports: a Method Description},
  booktitle = {Proceedings  of  the  ICML/UAI/COLT 2008 Workshop on Machine Learning for Health-Care Applications; 2008 July 9: Helsinki, Finland},
  author = {Suominen, Hanna and Ginter, Filip and Pyysalo, Sampo and Airola, Antti and Pahikkala, Tapio and Salanterä, Sanna and Salakoski, Tapio},
  editor = {Milos Hauskrecht, Dale Schuurmans Csaba Szepesvari},
  year = {2008},
}

Belongs to TUCS Research Unit(s): Turku BioNLP Group

Edit publication