Classifying social determinants of health from unstructured electronic health records using deep learning-based natural language processing

Han, Sifei; Zhang, Robert F.; Shi, Lei; Richie, Russell; Liu, Haixia; Tseng, Andrew; Quan, Wei; Ryan, Neal D.; Brent, David A.; Tsui, Fu-Chiang

doi:10.1016/j.jbi.2021.103984

Cited by 71 publications

(52 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, populational trend evaluations were performed for atrial fibrillation [ 58 ], left atrial appendage occlusion procedures [ 59 ], transcatheter aortic valve implantation and surgical aortic valve replacement operations [ 60 ], implantable cardioverter-defibrillators and cardiac resynchronization therapy [ 61 ]. Furthermore, NLP technology allows for the in-depth EHR assessment of social determinants, which are non-medical factors impacting patient health outcomes [ 62 , 63 , 64 ]. Leveraging this opportunity, AI systems can help to verify the correctness of the diagnoses, as well as provide valuable information on critical aspects associated with populational health.…”

Section: Discussionmentioning

confidence: 99%

Deep Learning Analysis of Polish Electronic Health Records for Diagnosis Prediction in Patients with Cardiovascular Diseases

Anetta

Horák

Wojakowski

et al. 2022

JPM

View full text Add to dashboard Cite

Electronic health records naturally contain most of the medical information in the form of doctor’s notes as unstructured or semi-structured texts. Current deep learning text analysis approaches allow researchers to reveal the inner semantics of text information and even identify hidden consequences that can offer extra decision support to doctors. In the presented article, we offer a new automated analysis of Polish summary texts of patient hospitalizations. The presented models were found to be able to predict the final diagnosis with almost 70% accuracy based just on the patient’s medical history (only 132 words on average), with possible accuracy increases when adding further sentences from hospitalization results; even one sentence was found to improve the results by 4%, and the best accuracy of 78% was achieved with five extra sentences. In addition to detailed descriptions of the data and methodology, we present an evaluation of the analysis using more than 50,000 Polish cardiology patient texts and dive into a detailed error analysis of the approach. The results indicate that the deep analysis of just the medical history summary can suggest the direction of diagnosis with a high probability that can be further increased just by supplementing the records with further examination results.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep Learning Analysis of Polish Electronic Health Records for Diagnosis Prediction in Patients with Cardiovascular Diseases

Anetta

Horák

Wojakowski

et al. 2022

JPM

View full text Add to dashboard Cite

show abstract

“…Among neural network architectures, feed-forward networks were only used in 3 studies (Table S11) [97] while BERT and variants were frequently used for phenotypes documented in clinical notes such as SDOHs (e.g. education [50,57]) and symptoms (e.g. chest pain, [90] bleeding [58]).…”

Section: Methodsmentioning

confidence: 99%

“…The phenotypes considered in articles utilizing traditional supervised learning were not identified in previous reviews [12,15] These include phenotypes primarily documented in free-text such as suicidal behavior [44,45] and SDOHs. [30,[46][47][48][49] Deep supervised learning papers similarly considered SDOHs [50][51][52][53][54][55][56][57] as well as episodic conditions [58][59][60][61] and COVID-19. [62,63] The phenotypes considered by articles using semi-or weakly-supervised methods aiming to expedite algorithm development included common, chronic conditions [64][65][66] that had been previously identified with a rule-based or traditional supervised learning method.…”

Section: Phenotypesmentioning

confidence: 99%

Machine Learning Approaches for Electronic Health Records Phenotyping: A Methodical Review

Yang

Varghese²,

Stephenson

et al. 2022

Preprint

View full text Add to dashboard Cite

ObjectiveAccurate and rapid methods for phenotyping are a prerequisite to realizing the potential of electronic health records (EHRs) data for clinical and translational research. This study reviews the literature on machine learning (ML) approaches for phenotyping with respect to the phenotypes considered, the data sources and methods used, and the contributions within the wider context of EHR-based research.Materials and MethodsWe searched for relevant articles in PubMed and Web of Science published between January 1, 2018 and April 14, 2022. After screening, we collected data on 52 variables across 106 selected articles.ResultsML-based methods were developed for 156 unique phenotypes, primarily using EHR data from a single institution or health system. 72 of 106 articles leveraged unstructured data in clinical notes. In terms of methodology, supervised learning is the most prevalent ML paradigm (n = 64, 60.4%), with half of the articles employing deep learning. Semi-supervised and weakly-supervised approaches were applied to reduce the burden of obtaining gold-standard labeled data (n = 21, 19.8%), while unsupervised learning was used for phenotype discovery (n = 20, 18.9%). Federated learning has been applied to develop algorithms across multiple institutions while preserving data privacy (n = 2, 1.9%).DiscussionWhile the use of ML for phenotyping is growing, most articles applied traditional supervised ML to characterize the presence of common, chronic conditions.ConclusionContinued research in ML-based methods is warranted, with particular attention to the development of advanced methods for complex phenotypes and standards for reporting and evaluating phenotyping algorithms.

show abstract

“…The typical drawback of physiological monitoring methods is the need for additional intrusive monitoring equipment or professional oversight, which puts the alternate strategy of performing screening using routinely collected electronic health records front and centre [11]. As PSG is expensive, time-consuming, and laborintensive, it is assumed that sleep physiological data likepulse oximetry as well as sleep stage length have significant prognostic capacity, but are not widely accessible [12,13]. Deep neural networks, which have many hidden layers instead of just one like ANN22-24, has wide range of applications in numerous fields [14,15].…”

Section: Contribution Of This Research Is As Followsmentioning

confidence: 99%

“…The features that are used to build the rule should be sparse or limited in the interim to make it simple to put into practise. The SVM border is, mathematically speaking, the answer to minimising by eq (13), 𝑄(𝛽, 𝑏, 𝜉) = (18) if we define SV as the set {j | αj > 0 for j = 1, 2, . .…”

Section: Weighted Curvature Based Feature Selection (Wcfs) With Suppo...mentioning

confidence: 99%

Computational Intelligence Based Electronic Healthcare Data Analytics Using Feature Selection with Classification by Deep Learning Architecture

Sivakumar¹

2022

ijfrcsce

View full text Add to dashboard Cite

EHRs (Electronic health records) are a source of big data that offer a wealth of clinical patient health data. However, because these notes are free-form texts, writing formats and styles range greatly amongst various records, text data from eHRs, such as discharge rapid notes, provide analysis challenges. This research proposed novel technique in electronic healthcare data analysis based on feature selection and classification utilizingDL methods. here the input is collected as input EH data, is processed for dimensionality reduction, noise removal. A public data pre-processing method for dealing with HD-EHR data is dimensionality reduction, which tries to minimize amount of EHR representational features while enhancing effectiveness of following data analysis, such as classification. The processed data features has been selected utilizingweighted curvature based feature selection with support vector machine. Then this selected deep features has been classified using sparse encoder transfer learning. the experimental analysis has been carried out for various EH datasets in terms of accuracy of 96%, precision of 92%, recall of 77%, F-1 score of 72%, MAP of 65%

show abstract

Classifying social determinants of health from unstructured electronic health records using deep learning-based natural language processing

Cited by 71 publications

References 28 publications

Deep Learning Analysis of Polish Electronic Health Records for Diagnosis Prediction in Patients with Cardiovascular Diseases

Deep Learning Analysis of Polish Electronic Health Records for Diagnosis Prediction in Patients with Cardiovascular Diseases

Machine Learning Approaches for Electronic Health Records Phenotyping: A Methodical Review

Computational Intelligence Based Electronic Healthcare Data Analytics Using Feature Selection with Classification by Deep Learning Architecture

Contact Info

Product

Resources

About