A Deep Learning Method for ICD-10 Coding of Free-Text Death Certificates

Duarte, Francisco; Martins, Bruno; Pinto, Cátia Sousa; Silva, Mário J.

doi:10.1007/978-3-319-65340-2_12

Cited by 12 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Five, we update the existing manual coding system based on the rule base of regexps to reduce workload and improve the work quality of coders. The technical requirements and computational cost are less than those of the other methods found in most studies [7,11], [32][33][34][35][36]. CNN [18,[34][35][36] is one of the state of the art proposals to solve the problem of automatic ICD coding.…”

Section: Discussionmentioning

confidence: 99%

“…For example, several studies based on machine learning approaches, such as the support vector machine (SVM) method [5][6][7][8], were proposed to automatically assign ICD-10 codes. With the extensive application of deep learning methods in various fields, these methods have also been widely used in automated ICD coding [9][10][11][12]. These studies indicate that deep learning models can produce interpretable results and can code automatically in a reasonable way.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Construction of a semi-automatic ICD-10 coding system

Zhou

Cheng

et al. 2020

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

Background: The International Classification of Diseases, 10th Revision (ICD-10) has been widely used to describe the diagnosis information of patients. Automatic ICD-10 coding is important because manually assigning codes is expensive, time consuming and error prone. Although numerous approaches have been developed to explore automatic coding, few of them have been applied in practice. Our aim is to construct a practical, automatic ICD-10 coding machine to improve coding efficiency and quality in daily work. Methods: In this study, we propose the use of regular expressions (regexps) to establish a correspondence between diagnosis codes and diagnosis descriptions in outpatient settings and at admission and discharge. The description models of the regexps were embedded in our upgraded coding system, which queries a diagnosis description and assigns a unique diagnosis code. Like most studies, the precision (P), recall (R), F-measure (F) and overall accuracy (A) were used to evaluate the system performance. Our study had two stages. The datasets were obtained from the diagnosis information on the homepage of the discharge medical record. The testing sets were from Results: The values of P were 89.27 and 88.38% in the first testing phase and the second testing phase, respectively, which demonstrate high precision. The automatic ICD-10 coding system completed more than 160,000 codes in 16 months, which reduced the workload of the coders. In addition, a comparison between the amount of time needed for manual coding and automatic coding indicated the effectiveness of the system-the time needed for automatic coding takes nearly 100 times less than manual coding. Conclusions: Our automatic coding system is well suited for the coding task. Further studies are warranted to perfect the description models of the regexps and to develop synthetic approaches to improve system performance.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Construction of a semi-automatic ICD-10 coding system

Zhou

Cheng

et al. 2020

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

show abstract

“…(1) The number of words in common between (a) the headline and the body of the news article, and (b) the headline and the first two sentences of the body; (2) Refutation features, based on the presence of refuting words, listed in a given dictionary, in the headline (e.g., words like deny, doubt, fraud or debunk); the headline and the first two sentences of the body; (6) The soft cosine similarity metric [6], computed between representations leveraging word occurrences for (a) the headline and the entire body, or (b) the headline and the first two sentences of the body; (7) The BLEU score [33] computed between (a) the headline and the set of sentences from the body, and (b) the headline and the first two sentences of the body; (8) ROUGE scores [27] computed between (a) the headline and the set of sentences from the body, and (b) the headline and the first two sentences of the body; (9) The CIDEr similarity score [47], computed between (a) the headline and the set of sentences of the article, and (b) the headline and the first two sentences of the body; (10) The cosine similarity metric, computed between TF-IDF vector representations for the words occurring in the headline, and in the body of the news article. (11) A vector representation of the headline, with 50 dimensions, resulting from a Singular Value Decomposition (SVD) of a matrix with TF-IDF representations for the texts; (12) A vector representation for the body of the article, with 50 dimensions, resulting from a Singular Value Decomposition (SVD) of a matrix with TF-IDF representations for the texts; (13) The cosine similarity metric, computed between the SVD vectors for the headline and the body; (14) A vector representation for the headline, with 300 dimensions, produced by averaging the word2vec embeddings for the words occurring in the headline; (15) A vector representation for the body, with 300 dimensions, produced by averaging the word2vec embeddings for all the words occurring in the body of the article; (16) The cosine similarity metric, computed between averaged word2vec embeddings for the headline and body; (17) Sentiment polarity scores for the headline and the body, computed with basis on a word polarity lexicon. The first 5 features from the previous enumeration were taken from the official FNC-1 baseline system, provided by the organizers.…”

Section: Combining the Matched Representations With External Featuresmentioning

confidence: 99%

“…• The proposed neural network architecture leverages a hierarchical approach for modeling the body of news articles, taking inspiration on previous studies addressing the classification of long documents [14,52]. In this approach, a Recurrent Neural Network (RNN) is used for modeling the sequence of sentences, which in turn are individually modeled by a separate RNN encoding sequences of words.…”

Section: Introductionmentioning

confidence: 99%

Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News

Borges

Martins

Calado

2019

J. Data and Information Quality

Self Cite

View full text Add to dashboard Cite

Fake news are nowadays an issue of pressing concern, given their recent rise as a potential threat to high-quality journalism and well-informed public discourse. The Fake News Challenge (FNC-1) was organized in early 2017 to encourage the development of machine learning-based classification systems for stance detection (i.e., for identifying whether a particular news article agrees, disagrees, discusses, or is unrelated to a particular news headline), thus helping in the detection and analysis of possible instances of fake news. This article presents a novel approach to tackle this stance detection problem, based on the combination of string similarity features with a deep neural network architecture that leverages ideas previously advanced in the context of learning efficient text representations, document classification, and natural language inference. Specifically, we use bi-directional Recurrent Neural Networks (RNNs), together with max-pooling over the temporal/sequential dimension and neural attention, for representing (i) the headline, (ii) the first two sentences of the news article, and (iii) the entire news article. These representations are then combined/compared, complemented with similarity features inspired on other FNC-1 approaches, and passed to a final layer that predicts the stance of the article towards the headline. We also explore the use of external sources of information, specifically large datasets of sentence pairs originally proposed for training and evaluating natural language inference methods, in order to pre-train specific components of the neural network architecture (e.g., the RNNs used for encoding sentences). The obtained results attest to the effectiveness of the proposed ideas and show that our model, particularly when considering pre-training and the combination of neural representations together with similarity features, slightly outperforms the previous state-of-the-art. 39:2 • Borges et al.is increasingly harder to know for sure what to trust, with the absorption of fake news by the masses having increasingly harmful consequences [48]. Automatically dealing with fake news has drawn considerable attention in several research communities [24,26,34,36,40,41,45]. However, the task of evaluating the veracity of news articles is still very demanding and complex, even for trained specialists and much more for automated systems.A useful first step towards identifying fake news articles relates to understanding what other news agencies, in a given moment, are reporting about the same topic. This sub-task is often referred to as stance detection, and automating this process might be useful in developing automated assistants to help in fact checking. In particular, an automatic approach to stance detection would allow, for example, someone to insert an allegation or a news title, and recover related articles that either agree, disagree, or discuss that title. Then, the human checker would use her own judgment to assess the situation.Based on the aforementioned general ideas, a Fake Ne...

show abstract

“…Duarte et al (2017) use hierarchical recurrent neural networks (RNNs) to annotate death reports with ICD–10 codes. Vani et al (2017) introduced grounded RNNs for EMR coding.…”

Section: Related Workmentioning

confidence: 99%

EMR Coding with Semi-Parametric Multi-Head Matching Networks

Rios

Kavuluru

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

View full text Add to dashboard Cite

Coding EMRs with diagnosis and procedure codes is an indispensable task for billing, secondary data analyses, and monitoring health trends. Both speed and accuracy of coding are critical. While coding errors could lead to more patient–side financial burden and mis–interpretation of a patient’s well–being, timely coding is also needed to avoid backlogs and additional costs for the healthcare facility. In this paper, we present a new neural network architecture that combines ideas from few–shot learning matching networks, multi–label loss functions, and convolutional neural networks for text classification to significantly outperform other state–of–the–art models. Our evaluations are conducted using a well known deidentified EMR dataset (MIMIC) with a variety of multi–label performance measures.

show abstract

A Deep Learning Method for ICD-10 Coding of Free-Text Death Certificates

Cited by 12 publications

References 10 publications

Construction of a semi-automatic ICD-10 coding system

Construction of a semi-automatic ICD-10 coding system

Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News

EMR Coding with Semi-Parametric Multi-Head Matching Networks

Contact Info

Product

Resources

About