Construction of a Semi-automatic ICD-10 Coding System

Zhou, Lingling; Cheng, Cheng; Ou, Dong; Hao, Huang

doi:10.21203/rs.3.rs-19839/v1

Cited by 5 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is difficult to compare our results with the literature since the problem definition and the finality are not always the same. For instance in terms of F1-Score, [7] obtained 46% for Spanish text classification, [4] reached 74% in hematology unit with 30 diagnosis, [3] attained 88% in radiology with 45 diagnosis, [8] obtained 76% with 6 diagnosis, [9] reached 46% with 500 diagnosis while in our work, we obtained 83% with 346 diagnosis.…”

Section: Discussionsupporting

confidence: 53%

“…[8] considered 6 principal diseases and proposed an unsupervised knowledge integration algorithm to analyze clinical narrative notes via semantic relevance assessment. [9] conceived a semi-automatic ICD-10 coding system based on regular expressions with promising results. The aim of this work is to assist hospitals in coding their stays by proposing an intelligent system based on NLP and DL approaches to automatically and accurately translates the free-text diagnosis descriptions into ICD codes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Deep Learning Framework for Automated ICD-10 Coding

Chraibi¹,

Delerue²,

Taillard³

et al. 2021

Studies in Health Technology and Informatics

View full text Add to dashboard Cite

The International Statistical Classification of Diseases and Related Health Problems (ICD) is one of the widely used classification system for diagnoses and procedures to assign diagnosis codes to Electronic Health Record (EHR) associated with a patient’s stay. The aim of this paper is to propose an automated coding system to assist physicians in the assignment of ICD codes to EHR. For this purpose, we created a pipeline of Natural Language Processing (NLP) and Deep Learning (DL) models able to extract the useful information from French medical texts and to perform classification. After the evaluation phase, our approach was able to predict 346 diagnosis codes from heterogeneous medical units with an accuracy average of 83%. Our results were finally validated by physicians of the Medical Information Department (MID) in charge of coding hospital stays.

show abstract

Section: Discussionsupporting

confidence: 53%

Section: Introductionmentioning

confidence: 99%

A Deep Learning Framework for Automated ICD-10 Coding

Chraibi¹,

Delerue²,

Taillard³

et al. 2021

Studies in Health Technology and Informatics

View full text Add to dashboard Cite

show abstract

“…Accuracy in providing diagnosis codes is an important thing that medical recorders must consider because the quality of coded data is vital for Health Information Management personnel. The accuracy of diagnostic data is crucial in clinical data management, cost collection, and other matters related to health care and services [29].…”

Section: Discussionmentioning

confidence: 99%

Synchronization of Codification of Unspecified Schizophrenia Againts Back-Referral System of Mirit Public Health Center

Wang¹,

Afriyani²,

Wahyuni³

et al. 2021

Dis.Prev.Public Heal J.

View full text Add to dashboard Cite

Background: The disease classification system is a grouping of diseases following the International Statistical Classification of Diseases and Related Health Problems Tenth Revisions ICD-10. The coding application must be in accordance with ICD-10 to obtain a valid code in disease indexing, national, international reporting of morbidity and mortality, analysis of health care costs, and epidemiological and clinical research. The diagnosis of schizoaffective disorder is made if schizophrenia and affective disorder are symptoms based on the ICD-10 diagnostic criteria. This study aims to determine the synchronization of the codification of unspecified schizophrenia and determine the factors that influence it against the back-referral system at the Mirit Health Center. Method: This research is qualitative research with a descriptive approach. Respondents were four officers, i.e. one doctor, one medical record officer, one person holding a mental program, one pharmacy officer. The number of observed medical record documents was 96 data with research indicators of accuracy and completeness of the diagnosis code in patients referred from First Level Health Facilities (FKTP) to Advanced Health Facilities (FKTL). Result: The results showed 30 referrals, with nine referrals having the accuracy of the patient referral diagnosis code. The back-referral program (PRB) for mental illness at the Mirit Health Center, in collaboration with Mbah Marsio's mental health rehabilitation center, was carried out well. However, the implementation of the Chronic Disease Management program (Prolanis) for mental health was still not good. Conclusion: There are many unsynchronized codifications. It is recommended to conduct an evaluation where they communicate with each other about the patient's condition to supervise the implementation of Referback Patients, especially to specialists who write the back-referral form.

show abstract

“…Identifying and organizing the rich information on individual function currently locked away in medical free text can unlock valuable details to enrich researchers' understanding of rehabilitation outcomes, and highlight salient details of patients' experiences in clinical decision making. Prior research on automated and semi-automated ICD coding systems using NLP methods provides an instructive example of how these approaches can streamline medical coding processes (36)(37)(38). Growing integration of the ICF into clinical and research settings, from primary care (39) and EHR implementation (40) to pediatric research (41), present similar opportunities to smooth the adoption and practical use of ICF categories with NLP-based coding systems.…”

Section: Broader Implications Of Icf Coding With Nlpmentioning

confidence: 99%

Linking Free Text Documentation of Functioning and Disability to the ICF with Natural Language Processing

Newman-Griffis

Maldonado

et al. 2021

Preprint

View full text Add to dashboard Cite

BackgroundInvaluable information on patient functioning and the complex interactions that define it is recorded in free text portions of the Electronic Health Record (EHR). Leveraging this information to improve clinical decision-making and conduct research requires natural language processing (NLP) technologies to identify and organize the information recorded in clinical documentation.MethodsWe used NLP methods to analyze information about patient functioning recorded in two collections of clinical documents pertaining to claims for federal disability benefits from the U.S. Social Security Administration (SSA). We grounded our analysis in the International Classification of Functioning, Disability and Health (ICF), and used the ICF’s Activities and Participation domain to classify information about functioning in three key areas: Mobility, Self-Care, and Domestic Life. After annotating functional status information in our datasets through expert clinical review, we trained machine learning-based NLP models to automatically assign ICF codes to mentions of functional activity.ResultsWe found that rich and diverse information on patient functioning was documented in the free text records. Annotation of 289 documents for Mobility information yielded 2,455 mentions of Mobility activities and 3,176 specific actions corresponding to 13 ICF-based codes. Annotation of 329 documents for Self-Care and Domestic Life information yielded 3,990 activity mentions and 4,665 specific actions corresponding to 16 ICF-based codes. NLP systems for automated ICF coding achieved over 80% macro-averaged F-measure on both datasets, indicating strong performance across all ICF codes used.ConclusionsNLP can help to navigate the tradeoff between flexible and expressive clinical documentation of functioning and standardizable data for comparability and learning. The ICF has practical limitations for classifying functional status information in clinical documentation, but presents a valuable framework for organizing the information recorded in health records about patient functioning. This study advances the development of robust, ICF-based NLP technologies to analyze information on patient functioning, and has significant implications for NLP-powered analysis of functional status information in disability benefits management, clinical care, and research.

show abstract

Construction of a Semi-automatic ICD-10 Coding System

Cited by 5 publications

References 13 publications

A Deep Learning Framework for Automated ICD-10 Coding

A Deep Learning Framework for Automated ICD-10 Coding

Synchronization of Codification of Unspecified Schizophrenia Againts Back-Referral System of Mirit Public Health Center

Linking Free Text Documentation of Functioning and Disability to the ICF with Natural Language Processing

Contact Info

Product

Resources

About