other · NORMA · 2022

ICD-10 Code Prediction using Machine Learning

Samynathan, Sarath Kumar

Cited by 0Abstract and details only

Abstract

Unstructured data such as free text available in Electronic Health Records used in medical organizations are complex and hard to handle manually. Machine learning can be used to convert free text into vectors of tokens with the help of natural language processing and predict the outcome based on such conversion. Our aim of this project was to predict the ICD codes based on the synonyms of the diseases and we have predicted with the help of various machine learning algorithms and neural networks. We have used Random Forest, Support Vector, Logistic regression, Naive Bayes, KNN and MLP classifier to predict the ICD codes based on synonyms. We have also used the multilanguage embedding LASER so that the work can be used in multiple languages. The algorithm with highest accuracy is a random forest classifier, achieving 98% accuracy. Our results show that a reliable way to search for medical synonyms is made possible with traditional machine learning techniques. We therefore think that any user of such an application can successfully predict ICD codes based on synonyms.

Cite this paper

APA 7

Samynathan, S. K. (2022). ICD-10 Code Prediction using Machine Learning. NORMA.

MLA 9

Samynathan, Sarath Kumar. “ICD-10 Code Prediction using Machine Learning.” NORMA, 2022.

Chicago

Samynathan, Sarath Kumar. 2022. “ICD-10 Code Prediction using Machine Learning.” NORMA.

Harvard

Samynathan, S.K. (2022) ‘ICD-10 Code Prediction using Machine Learning’, NORMA.

BibTeX

@misc{samynathan2022code,
  title = {{ICD-10 Code Prediction using Machine Learning}},
  author = {Samynathan, Sarath Kumar},
  journal = {NORMA},
  year = {2022}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.