ICD-9 Tagging of Clinical Notes Using Topical Word Embedding

Samonte, Mary Jane C.; Gerardo, Bobby D.; Fajardo, Arnel C.; Medina, Ruji P.

doi:10.1145/3230348.3230357

Cited by 9 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The majority of studies have applied tokenization, followed by removal of stop words, removal of non-alphabetic characters and lowercase conversion. Apart from that, few studies also used other data processing steps such as regular expression matching [88], building dictionary or vocabulary [5,52], removing non-matching terms [78], removal of de-identified or confidential information [74,36]. Studies including [67,5,61,90,84,50,58,37] truncated the documents to a maximum length of 2,500 or 4,000 tokens in order to reduce the computational cost.…”

Section: Preprocessingmentioning

confidence: 99%

“…To summarise, the majority of studies (n=17) applied Word2Vec embedding, followed by TF-IDF feature representation and BoW. Few studies Ayyar and Oliver [4], Lin et al [52], and Mascio et al [56] used GloVe embeddings, while Samonte et al [74] applied topical word embedding. There are a few studies that have not reported the embedding model except dimensions of embedding.…”

Section: Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Literature Review of Automated ICD Coding and Classification Systems using Discharge Summaries

Kaur,

Ginige,

Obst

2021

Preprint

View full text Add to dashboard Cite

A systematic literature review focus on automated ICD code assignment using discharge summaries was conducted.• A total of 38 studies published between January 2010 and December 2020 were selected and analysed.• We review computerised systems that employ Artificial Intelligence, Machine Learning, Deep Learning and Natural Language Processing for assigning ICD codes to discharge summaries.• We highlighted limitations in existing studies and discussed open research challenges.

show abstract

Section: Preprocessingmentioning

confidence: 99%

Section: Feature Extractionmentioning

confidence: 99%

A Systematic Literature Review of Automated ICD Coding and Classification Systems using Discharge Summaries

Kaur,

Ginige,

Obst

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Clinical notes are used to classify the top 10 ICD-9 codes and blocks. Enhanced Hierarchical Attention Network (EnHAN) [24] utilizes discharge summary to solve ICD-9 prediction problems. To deal with multi-class label problems, the method uses topical word embedding.…”

Section: Related Workmentioning

confidence: 99%

Predicting ICD-9 Codes Using Self-Report of Patients

Singaravelan

Hsieh²,

Liao

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

The International Classification of Diseases (ICD) is a globally recognized medical classification system that aids in the identification of diseases and the regulation of health trends. The ICD framework makes it easy to keep track of records and evaluate medical data for evidence-based decision-making. Several methods have predicted ICD-9 codes based on the discharge summary, clinical notes, and nursing notes. In our study, our approach only utilizes the subjective component to predict ICD-9 codes. Data cleaning and segmentation, and Natural Language Processing (NLP) techniques are applied on the subjective component during the pre-processing. Our study builds the Long Short-Term Memory (LSTM) and the Gated Recurrent Unit (GRU) to develop a model for predicting ICD-9 codes. The ICD-9 codes contain different ICD levels such as chapter, block, three-digit code, and full code. The GRU model scores the highest recall of 57.91% in the chapter level and the top-10 experiment has a recall of 67.37%. Based on the subjective component, the model can help patients in the form of a remote assistance tool.

show abstract

“…Patients may have more than one medical problems, which leads to multiple medical diagnoses. In the study by Samonte et al [ 20 ], the proposed method of Enhanced Hierarchical Attention Network (EnHAN) followed topical word embedding and word embedding to solve this multi-class labeling and multi-label classification approach. This approach achieved a high accuracy of 0.841.…”

Section: Introductionmentioning

confidence: 99%

“…Note that the authors in [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 ] used patient profile, clinical examination reports and physician diagnosis results to establish models for the prediction of certain diseases that are commonly seen or with high mortality rates. On the contrary, in this paper, we only make use of patients’ self-report data (i.e., the subjective component in the progress note of EMR) to establish a predictive model for a variety of diseases.…”

Section: Introductionmentioning

confidence: 99%

Applying Convolutional Neural Networks to Predict the ICD-9 Codes of Medical Records

Hsu

Hsieh³

et al. 2020

Sensors

View full text Add to dashboard Cite

The International Statistical Classification of Disease and Related Health Problems (ICD) is an international standard system for categorizing and reporting diseases, injuries, disorders, and health conditions. Most previously-proposed disease predicting systems need clinical information collected by the medical staff from the patients in hospitals. In this paper, we propose a deep learning algorithm to classify disease types and identify diagnostic codes by using only the subjective component of progress notes in medical records. In this study, we have a dataset, consisting of about one hundred and sixty-eight thousand medical records, from a medical center, collected during 2003 and 2017. First, we apply standard text processing procedures to parse the sentences and word embedding techniques for vector representations. Next, we build a convolution neural network model on the medical records to predict the ICD-9 code by using a subjective component of the progress note. The prediction performance is evaluated by ten-fold cross-validation and yields an accuracy of 0.409, recall of 0.409 and precision of 0.436. If we only consider the “chapter match” of ICD-9 code, our model achieves an accuracy of 0.580, recall of 0.580, and precision of 0.582. Since our diagnostic code prediction model is solely based on subjective components (mainly, patients’ self-report descriptions), the proposed approach could serve as a remote and self-diagnosis assistance tool, prior to seeking medical advice or going to the hospital. In addition, our work may be used as a primary evaluation tool for discomfort in the rural area where medical resources are restricted.

show abstract

ICD-9 Tagging of Clinical Notes Using Topical Word Embedding

Cited by 9 publications

References 23 publications

A Systematic Literature Review of Automated ICD Coding and Classification Systems using Discharge Summaries

A Systematic Literature Review of Automated ICD Coding and Classification Systems using Discharge Summaries

Predicting ICD-9 Codes Using Self-Report of Patients

Applying Convolutional Neural Networks to Predict the ICD-9 Codes of Medical Records

Contact Info

Product

Resources

About