Extracting Family History of Patients From Clinical Narratives: Exploring an End-to-End Solution With Deep Learning Models

Deep Learning for Combating Misinformation in Multicategorical Text Contents

Currently, one can observe the evolution of social media networks. In particular, humans are faced with the fact that, often, the opinion of an expert is as important and significant as the opinion of a non-expert. It is possible to observe changes and processes in traditional media that reduce the role of a conventional 'editorial office', placing gradual emphasis on the remote work of journalists and forcing increasingly frequent use of online sources rather than actual reporting work. As a result, social media has become an element of state security, as disinformation and fake news produced by malicious actors can manipulate readers, creating unnecessary debate on topics organically irrelevant to society. This causes a cascading effect, fear of citizens, and eventually threats to the state's security. Advanced data sensors and deep machine learning methods have great potential to enable the creation of effective tools for combating the fake news problem. However, these solutions often need better model generalization in the real world due to data deficits. In this paper, we propose an innovative solution involving a committee of classifiers in order to tackle the fake news detection challenge. In that regard, we introduce a diverse set of base models, each independently trained on sub-corpora with unique characteristics. In particular, we use multi-label text category classification, which helps formulate an ensemble. The experiments were conducted on six different benchmark datasets. The results are promising and open the field for further research.

Clinical Concept-Based Radiology Reports Classification Pipeline for Lung Carcinoma

Rising incidence and mortality of cancer have led to an incremental amount of research in the field. To learn from preexisting data, it has become important to capture maximum information related to disease type, stage, treatment, and outcomes. Medical imaging reports are rich in this kind of information but are only present as free text. The extraction of information from such unstructured text reports is labor-intensive. The use of Natural Language Processing (NLP) tools to extract information from radiology reports can make it less time-consuming as well as more effective. In this study, we have developed and compared different models for the classification of lung carcinoma reports using clinical concepts. This study was approved by the institutional ethics committee as a retrospective study with a waiver of informed consent. A clinical concept-based classification pipeline for lung carcinoma radiology reports was developed using rule-based as well as machine learning models and compared. The machine learning models used were XGBoost and two more deep learning model architectures with bidirectional long short-term neural networks. A corpus consisting of 1700 radiology reports including computed tomography (CT) and positron emission tomography/computed tomography (PET/CT) reports were used for development and testing. Five hundred one radiology reports from MIMIC-III Clinical Database version 1.4 was used for external validation. The pipeline achieved an overall F1 score of 0.94 on the internal set and 0.74 on external validation with the rule-based algorithm using expert input giving the best performance. Among the machine learning models, the Bi-LSTM_dropout model performed better than the ML model using XGBoost and the Bi-LSTM_simple model on internal set, whereas on external validation, the Bi-LSTM_simple model performed relatively better than other 2. This pipeline can be used for clinical concept-based classification of radiology reports related to lung carcinoma from a huge corpus and also for automated annotation of these reports.

Information extraction from electronic medical documents: state of the art and future research directions

In the medical field, a doctor must have a comprehensive knowledge by reading and writing narrative documents, and he is responsible for every decision he takes for patients. Unfortunately, it is very tiring to read all necessary information about drugs, diseases and patients due to the large amount of documents that are increasing every day. Consequently, so many medical errors can happen and even kill people. Likewise, there is such an important field that can handle this problem, which is the information extraction. There are several important tasks in this field to extract the important and desired information from unstructured text written in natural language. The main principal tasks are named entity recognition and relation extraction since they can structure the text by extracting the relevant information. However, in order to treat the narrative text we should use natural language processing techniques to extract useful information and features. In our paper, we introduce and discuss the several techniques and solutions used in these tasks. Furthermore, we outline the challenges in information extraction from medical documents. In our knowledge, this is the most comprehensive survey in the literature with an experimental analysis and a suggestion for some uncovered directions.

A large language model for electronic health records

There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. However, there are few clinical language models, the largest of which trained in the clinical domain is comparatively small at 110 million parameters (compared with billions of parameters in the general domain). It is not clear how large clinical language models with billions of parameters can help medical AI systems utilize unstructured EHRs. In this study, we develop from scratch a large clinical language model-GatorTron-using >90 billion words of text (including >82 billion words of de-identified clinical text) and systematically evaluate it on five clinical NLP tasks including clinical concept extraction, medical relation extraction, semantic textual similarity, natural language inference (NLI), and medical question answering (MQA). We examine how (1) scaling up the number of parameters and (2) scaling up the size of the training data could benefit these NLP tasks. GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks (e.g., 9.6% and 9.5% improvement in accuracy for NLI and MQA), which can be applied to medical AI systems to improve healthcare delivery. The GatorTron models are publicly available at: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/models/gatortron_og .

Identify diabetic retinopathy-related clinical concepts and their attributes using transformer-based natural language processing methods

BACKGROUND

Diabetic retinopathy (DR) is a leading cause of blindness in American adults. If detected, DR can be treated to prevent further damage causing blindness. There is an increasing interest in developing artificial intelligence (AI) technologies to help detect DR using electronic health records. The lesion-related information documented in fundus image reports is a valuable resource that could help diagnoses of DR in clinical decision support systems. However, most studies for AI-based DR diagnoses are mainly based on medical images; there is limited studies to explore the lesion-related information captured in the free text image reports.

METHODS

In this study, we examined two state-of-the-art transformer-based natural language processing (NLP) models, including BERT and RoBERTa, compared them with a recurrent neural network implemented using Long short-term memory (LSTM) to extract DR-related concepts from clinical narratives. We identified four different categories of DR-related clinical concepts including lesions, eye parts, laterality, and severity, developed annotation guidelines, annotated a DR-corpus of 536 image reports, and developed transformer-based NLP models for clinical concept extraction and relation extraction. We also examined the relation extraction under two settings including 'gold-standard' setting-where gold-standard concepts were used-and end-to-end setting.

RESULTS

For concept extraction, the BERT model pretrained with the MIMIC III dataset achieve the best performance (0.9503 and 0.9645 for strict/lenient evaluation). For relation extraction, BERT model pretrained using general English text achieved the best strict/lenient F1-score of 0.9316. The end-to-end system, BERT_general_e2e, achieved the best strict/lenient F1-score of 0.8578 and 0.8881, respectively. Another end-to-end system based on the RoBERTa architecture, RoBERTa_general_e2e, also achieved the same performance as BERT_general_e2e in strict scores.

CONCLUSIONS

This study demonstrated the efficiency of transformer-based NLP models for clinical concept extraction and relation extraction. Our results show that it's necessary to pretrain transformer models using clinical text to optimize the performance for clinical concept extraction. Whereas, for relation extraction, transformers pretrained using general English text perform better.

Identifying Patients Who Meet Criteria for Genetic Testing of Hereditary Cancers Based on Structured and Unstructured Family Health History Data in the Electronic Health Record: Natural Language Processing Approach

BACKGROUND

Family health history has been recognized as an essential factor for cancer risk assessment and is an integral part of many cancer screening guidelines, including genetic testing for personalized clinical management strategies. However, manually identifying eligible candidates for genetic testing is labor intensive.

OBJECTIVE

The aim of this study was to develop a natural language processing (NLP) pipeline and assess its contribution to identifying patients who meet genetic testing criteria for hereditary cancers based on family health history data in the electronic health record (EHR). We compared an algorithm that uses structured data alone with structured data augmented using NLP.

METHODS

Algorithms were developed based on the National Comprehensive Cancer Network (NCCN) guidelines for genetic testing for hereditary breast, ovarian, pancreatic, and colorectal cancers. The NLP-augmented algorithm uses both structured family health history data and the associated unstructured free-text comments. The algorithms were compared with a reference standard of 100 patients with a family health history in the EHR.

RESULTS

Regarding identifying the reference standard patients meeting the NCCN criteria, the NLP-augmented algorithm compared with the structured data algorithm yielded a significantly higher recall of 0.95 (95% CI 0.9-0.99) versus 0.29 (95% CI 0.19-0.40) and a precision of 0.99 (95% CI 0.96-1.00) versus 0.81 (95% CI 0.65-0.95). On the whole data set, the NLP-augmented algorithm extracted 33.6% more entities, resulting in 53.8% more patients meeting the NCCN criteria.

CONCLUSIONS

Compared with the structured data algorithm, the NLP-augmented algorithm based on both structured and unstructured family health history data in the EHR increased the number of patients identified as meeting the NCCN criteria for genetic testing for hereditary breast or ovarian and colorectal cancers.

The Value of Extracting Clinician-Recorded Affect for Advancing Clinical Research on Depression: Proof-of-Concept Study Applying Natural Language Processing to Electronic Health Records

Background

Affective characteristics are associated with depression severity, course, and prognosis. Patients’ affect captured by clinicians during sessions may provide a rich source of information that more naturally aligns with the depression course and patient-desired depression outcomes.

Objective

In this paper, we propose an information extraction vocabulary used to pilot the feasibility and reliability of identifying clinician-recorded patient affective states in clinical notes from electronic health records.

Methods

Affect and mood were annotated in 147 clinical notes of 109 patients by 2 independent coders across 3 pilots. Intercoder discrepancies were settled by a third coder. This reference annotation set was used to test a proof-of-concept natural language processing (NLP) system using a named entity recognition approach.

Results

Concepts were frequently addressed in templated format and free text in clinical notes. Annotated data demonstrated that affective characteristics were identified in 87.8% (129/147) of the notes, while mood was identified in 97.3% (143/147) of the notes. The intercoder reliability was consistently good across the pilots (interannotator agreement [IAA] >70%). The final NLP system showed good reliability with the final reference annotation set (mood IAA=85.8%; affect IAA=80.9%).

Conclusions

Affect and mood can be reliably identified in clinician reports and are good targets for NLP. We discuss several next steps to expand on this proof of concept and the value of this research for depression clinical research.

Ambiguous and Incomplete: Natural Language Processing Reveals Problematic Reporting Styles in Thyroid Ultrasound Reports

OBJECTIVE

 Natural language processing (NLP) systems convert unstructured text into analyzable data. Here, we describe the performance measures of NLP to capture granular details on nodules from thyroid ultrasound (US) reports and reveal critical issues with reporting language.

METHODS

 We iteratively developed NLP tools using clinical Text Analysis and Knowledge Extraction System (cTAKES) and thyroid US reports from 2007 to 2013. We incorporated nine nodule features for NLP extraction. Next, we evaluated the precision, recall, and accuracy of our NLP tools using a separate set of US reports from an academic medical center (A) and a regional health care system (B) during the same period. Two physicians manually annotated each test-set report. A third physician then adjudicated discrepancies. The adjudicated "gold standard" was then used to evaluate NLP performance on the test-set.

RESULTS

 A total of 243 thyroid US reports contained 6,405 data elements. Inter-annotator agreement for all elements was 91.3%. Compared with the gold standard, overall recall of the NLP tool was 90%. NLP recall for thyroid lobe or isthmus characteristics was: laterality 96% and size 95%. NLP accuracy for nodule characteristics was: laterality 92%, size 92%, calcifications 76%, vascularity 65%, echogenicity 62%, contents 76%, and borders 40%. NLP recall for presence or absence of lymphadenopathy was 61%. Reporting style accounted for 18% errors. For example, the word "heterogeneous" interchangeably referred to nodule contents or echogenicity. While nodule dimensions and laterality were often described, US reports only described contents, echogenicity, vascularity, calcifications, borders, and lymphadenopathy, 46, 41, 17, 15, 9, and 41% of the time, respectively. Most nodule characteristics were equally likely to be described at hospital A compared with hospital B.

CONCLUSIONS

 NLP can automate extraction of critical information from thyroid US reports. However, ambiguous and incomplete reporting language hinders performance of NLP systems regardless of institutional setting. Standardized or synoptic thyroid US reports could improve NLP performance.

A Survey on Recent Named Entity Recognition and Relationship Extraction Techniques on Clinical Texts

Significant growth in Electronic Health Records (EHR) over the last decade has provided an abundance of clinical text that is mostly unstructured and untapped. This huge amount of clinical text data has motivated the development of new information extraction and text mining techniques. Named Entity Recognition (NER) and Relationship Extraction (RE) are key components of information extraction tasks in the clinical domain. In this paper, we highlight the present status of clinical NER and RE techniques in detail by discussing the existing proposed NLP models for the two tasks and their performances and discuss the current challenges. Our comprehensive survey on clinical NER and RE encompass current challenges, state-of-the-art practices, and future directions in information extraction from clinical text. This is the first attempt to discuss both of these interrelated topics together in the clinical context. We identified many research articles published based on different approaches and looked at applications of these tasks. We also discuss the evaluation metrics that are used in the literature to measure the effectiveness of the two these NLP methods and future research directions.

A Study of Social and Behavioral Determinants of Health in Lung Cancer Patients Using Transformers-based Natural Language Processing Models

Social and behavioral determinants of health (SBDoH) have important roles in shaping people's health. In clinical research studies, especially comparative effectiveness studies, failure to adjust for SBDoH factors will potentially cause confounding issues and misclassification errors in either statistical analyses and machine learning-based models. However, there are limited studies to examine SBDoH factors in clinical outcomes due to the lack of structured SBDoH information in current electronic health record (EHR) systems, while much of the SBDoH information is documented in clinical narratives. Natural language processing (NLP) is thus the key technology to extract such information from unstructured clinical text. However, there is not a mature clinical NLP system focusing on SBDoH. In this study, we examined two state-of-the-art transformer-based NLP models, including BERT and RoBERTa, to extract SBDoH concepts from clinical narratives, applied the best performing model to extract SBDoH concepts on a lung cancer screening patient cohort, and examined the difference of SBDoH information between NLP extracted results and structured EHRs (SBDoH information captured in standard vocabularies such as the International Classification of Diseases codes). The experimental results show that the BERT-based NLP model achieved the best strict/lenient F1-score of 0.8791 and 0.8999, respectively. The comparison between NLP extracted SBDoH information and structured EHRs in the lung cancer patient cohort of 864 patients with 161,933 various types of clinical notes showed that much more detailed information about smoking, education, and employment were only captured in clinical narratives and that it is necessary to use both clinical narratives and structured EHRs to construct a more complete picture of patients' SBDoH factors.