Electronic case report forms generation from pathology reports by ARGO, automatic record generator for onco-hematology.
Ontology highlight
ABSTRACT:
SUBMITTER: Zaccaria GM
PROVIDER: S-EPMC8664934 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:ObjectivesElectronic Health Records (EHRs) contain scanned documents from a variety of sources such as identification cards, radiology reports, clinical correspondence, and many other document types. We describe the distribution of scanned documents at one health institution and describe the design and evaluation of a system to categorize documents into clinically relevant and non-clinically relevant categories as well as further sub-classifications. Our objective is to demonstrate that text classification systems can accurately classify scanned documents.MethodsWe extracted text using Optical Character Recognition (OCR). We then created and evaluated multiple text classification machine learning models, including both "bag of words" and deep learning approaches. We evaluated the system on three different levels of classification using both the entire document as input, as well as the individual pages of the document. Finally, we compared the effects of different text processing methods.ResultsA deep learning model using ClinicalBERT performed best. This model distinguished between clinically-relevant documents and not clinically-relevant documents with an accuracy of 0.973; between intermediate sub-classifications with an accuracy of 0.949; and between individual classes with an accuracy of 0.913.DiscussionWithin the EHR, some document categories such as "external medical records" may contain hundreds of scanned pages without clear document boundaries. Without further sub-classification, clinicians must view every page or risk missing clinically-relevant information. Machine learning can automatically classify these scanned documents to reduce clinician burden.ConclusionUsing machine learning applied to OCR-extracted text has the potential to accurately identify clinically-relevant scanned content within EHRs.
| S-EPMC7731898 | biostudies-literature
Project description:ObjectivePathology reports are rich in narrative statements that encode a complex web of relations among medical concepts. These relations are routinely used by doctors to reason on diagnoses, but often require hand-crafted rules or supervised learning to extract into prespecified forms for computational disease modeling. We aim to automatically capture relations from narrative text without supervision.MethodsWe design a novel framework that translates sentences into graph representations, automatically mines sentence subgraphs, reduces redundancy in mined subgraphs, and automatically generates subgraph features for subsequent classification tasks. To ensure meaningful interpretations over the sentence graphs, we use the Unified Medical Language System Metathesaurus to map token subsequences to concepts, and in turn sentence graph nodes. We test our system with multiple lymphoma classification tasks that together mimic the differential diagnosis by a pathologist. To this end, we prevent our classifiers from looking at explicit mentions or synonyms of lymphomas in the text.Results and conclusionsWe compare our system with three baseline classifiers using standard n-grams, full MetaMap concepts, and filtered MetaMap concepts. Our system achieves high F-measures on multiple binary classifications of lymphoma (Burkitt lymphoma, 0.8; diffuse large B-cell lymphoma, 0.909; follicular lymphoma, 0.84; Hodgkin lymphoma, 0.912). Significance tests show that our system outperforms all three baselines. Moreover, feature analysis identifies subgraph features that contribute to improved performance; these features agree with the state-of-the-art knowledge about lymphoma classification. We also highlight how these unsupervised relation features may provide meaningful insights into lymphoma classification.
| S-EPMC4147603 | biostudies-literature
Project description:Semantic interoperability of clinical data is essential to preserve its meaning and intent when the data is exchanged, re-used or integrated with other data. Achieving semantic operability requires the use of a communication standard, such as HL7, as well as (functional) information standards. Manually mapping clinical data to a medical thesaurus, such as SNOMED CT, is complicated and requires expert knowledge of both the dataset, including its context, and the thesaurus. As an alternative, the (re-)use of codebooks, data definitions which may already have been mapped to a thesaurus, can be a viable approach. We've developed the iCRF Generator, a Java program that can generate the core of an interoperable electronic case report form (iCRF) for several of the major electronic data capture systems (EDCs). To build their CRFs, users can select one or more items from established codebooks, available from an online system called ART-DECOR. By providing an easy to use method to create CRFs for multiple EDCs based on the same codebooks, interoperability can be more easily attained.
| S-EPMC7291075 | biostudies-literature
Project description:Background:Polycystic ovary syndrome (PCOS) is characterized by hyperandrogenemia, oligo-anovulation, and numerous ovarian cysts. Hospital electronic medical records provide an avenue for investigating polycystic ovary morphology commonly seen in PCOS at a large scale. The purpose of this study was to develop and evaluate the performance of two machine learning text algorithms, for classification of polycystic ovary morphology (PCOM) in pelvic ultrasounds. Methods:Pelvic ultrasound reports from patients at Boston Medical Center between October 1, 2003 and December 12, 2016 were included for analysis, which resulted in 39,093 ultrasound reports from 25,535 unique women. Following the 2003 Rotterdam Consensus Criteria for polycystic ovary syndrome, 2000 randomly selected ultrasounds were expert labeled for PCOM status as present, absent, or unidentifiable (not able to be determined from text alone). An ovary was marked as having PCOM if there was mention of numerous peripheral follicles or if the volume was greater than 10?ml in the absence of a dominant follicle or other confounding pathology. Half of the labeled data was used to develop and refine the algorithms, and the other half was used as a test set for evaluating its accuracy. Results:On the evaluation set of 1000 random US reports, the accuracy of the classifiers were 97.6% (95% CI: 96.5, 98.5%) and 96.1% (94.7, 97.2%). Both models were more adept at identifying PCOM-absent ultrasounds than either PCOM-unidentifiable or PCOM-present ultrasounds. The two classifiers estimated prevalence of PCOM within the whole set of 39,093 ultrasounds to be 44% PCOM-absent, 32% PCOM-unidentifiable, and 24% PCOM-present. Conclusions:Although accuracy measured on the test set and inter-rater agreement between the two classifiers (Cohen's Kappa?=?0.988) was high, a major limitation of our approach is that it uses the ultrasound report text as a proxy and does not directly count follicles from the ultrasound images themselves.
| S-EPMC6886196 | biostudies-literature
Project description:IntroductionElectronic health records (EHR) are linked together to examine disease history and to undertake research into the causes and outcomes of disease. However, the process of constructing algorithms for phenotyping (e.g., identifying disease characteristics) or health characteristics (e.g., smoker) is very time consuming and resource costly. In addition, results can vary greatly between researchers. Reusing or building on algorithms that others have created is a compelling solution to these problems. However, sharing algorithms is not a common practice and many published studies do not detail the clinical code lists used by the researchers in the disease/characteristic definition. To address these challenges, a number of centres across the world have developed health data portals which contain concept libraries (e.g., algorithms for defining concepts such as disease and characteristics) in order to facilitate disease phenotyping and health studies.ObjectivesThis study aims to review the literature of existing concept libraries, examine their utilities, identify the current gaps, and suggest future developments.MethodsThe five-stage framework of Arksey and O'Malley was used for the literature search. This approach included defining the research questions, identifying relevant studies through literature review, selecting eligible studies, charting and extracting data, and summarising and reporting the findings.ResultsThis review identified seven publicly accessible Electronic Health data concept libraries which were developed in different countries including UK, USA, and Canada. The concept libraries (n = 7) investigated were either general libraries that hold phenotypes of multiple specialties (n = 4) or specialized libraries that manage only certain specialities such as rare diseases (n = 3). There were some clear differences between the general libraries such as archiving data from different electronic sources, and using a range of different types of coding systems. However, they share some clear similarities such as enabling users to upload their own code lists, and allowing users to use/download the publicly accessible code. In addition, there were some differences between the specialized libraries such as difference in ability to search, and if it was possible to use different searching queries such as simple or complex searches. Conversely, there were some similarities between the specialized libraries such as enabling users to upload their own concepts into the libraries and to show where they were published, which facilitates assessing the validity of the concepts. All the specialized libraries aimed to encourage the reuse of research methods such as lists of clinical code and/or metadata.ConclusionThe seven libraries identified have been developed independently and appear to replicate similar concepts but in different ways. Collaboration between similar libraries would greatly facilitate the use of these libraries for the user. The process of building code lists takes time and effort. Access to existing code lists increases consistency and accuracy of definitions across studies. Concept library developers should collaborate with each other to raise awareness of their existence and of their various functions, which could increase users' contributions to those libraries and promote their wide-ranging adoption.
| S-EPMC8210840 | biostudies-literature
Project description:The upgraded knowledge of tumor biology and microenviroment provides information on differences in neoplastic and normal cells. Thus, the need to target these differences led to the development of novel molecules (targeted therapy) active against the neoplastic cells' inner workings. There are several types of targeted agents, including Small Molecules Inhibitors (SMIs), monoclonal antibodies (mAbs), interfering RNA (iRNA) molecules and microRNA. In the clinical practice, these new medicines generate a multilayered step in pharmacokinetics (PK), which encompasses a broad individual PK variability, and unpredictable outcomes according to the pharmacogenetics (PG) profile of the patient (e.g., cytochrome P450 enzyme), and to patient characteristics such as adherence to treatment and environmental factors. This review focuses on the use of targeted agents in-human phase I/II/III clinical trials in cancer-hematology. Thus, it outlines the up-to-date anticancer drugs suitable for targeted therapies and the most recent finding in pharmacogenomics related to drug response. Besides, a summary assessment of the genotyping costs has been discussed. Targeted therapy seems to be an effective and less toxic therapeutic approach in onco-hematology. The identification of individual PG profile should be a new resource for oncologists to make treatment decisions for the patients to minimize the toxicity and or inefficacy of therapy. This could allow the clinicians to evaluate benefits and restrictions, regarding costs and applicability, of the most suitable pharmacological approach for performing a tailor-made therapy.
| S-EPMC6723645 | biostudies-literature
Project description:BackgroundTo implement standardized machine-processable clinical sequencing reports in an electronic health record (EHR) system, the International Organization for Standardization Technical Specification (ISO/TS) 20428 international standard was proposed for a structured template. However, there are no standard implementation guidelines for data items from the proposed standard at the clinical site and no guidelines or references for implementing gene sequencing data results for clinical use. This is a significant challenge for implementation and application of these standards at individual sites.ObjectiveThis study examines the field utilization of genetic test reports by designing the Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) for genomic data elements based on the ISO/TS 20428 standard published as the standard for genomic test reports. The goal of this pilot is to facilitate the reporting and viewing of genomic data for clinical applications. FHIR Genomics resources predominantly focus on transmitting or representing sequencing data, which is of less clinical value.MethodsIn this study, we describe the practical implementation of ISO/TS 20428 using HL7 FHIR Genomics implementation guidance to efficiently deliver the required genomic sequencing results to clinicians through an EHR system.ResultsWe successfully administered a structured genomic sequencing report in a tertiary hospital in Korea based on international standards. In total, 90 FHIR resources were used. Among 41 resources for the required fields, 26 were reused and 15 were extended. For the optional fields, 28 were reused and 21 were extended.ConclusionsTo share and apply genomic sequencing data in both clinical practice and translational research, it is essential to identify the applicability of the standard-based information system in a practical setting. This prototyping work shows that reporting data from clinical genomics sequencing can be effectively implemented into an EHR system using the existing ISO/TS 20428 standard and FHIR resources.
| S-EPMC7445611 | biostudies-literature
Project description:ObjectiveWe implement 2 different multitask learning (MTL) techniques, hard parameter sharing and cross-stitch, to train a word-level convolutional neural network (CNN) specifically designed for automatic extraction of cancer data from unstructured text in pathology reports. We show the importance of learning related information extraction (IE) tasks leveraging shared representations across the tasks to achieve state-of-the-art performance in classification accuracy and computational efficiency.Materials and methodsMultitask CNN (MTCNN) attempts to tackle document information extraction by learning to extract multiple key cancer characteristics simultaneously. We trained our MTCNN to perform 5 information extraction tasks: (1) primary cancer site (65 classes), (2) laterality (4 classes), (3) behavior (3 classes), (4) histological type (63 classes), and (5) histological grade (5 classes). We evaluated the performance on a corpus of 95 231 pathology documents (71 223 unique tumors) obtained from the Louisiana Tumor Registry. We compared the performance of the MTCNN models against single-task CNN models and 2 traditional machine learning approaches, namely support vector machine (SVM) and random forest classifier (RFC).ResultsMTCNNs offered superior performance across all 5 tasks in terms of classification accuracy as compared with the other machine learning models. Based on retrospective evaluation, the hard parameter sharing and cross-stitch MTCNN models correctly classified 59.04% and 57.93% of the pathology reports respectively across all 5 tasks. The baseline models achieved 53.68% (CNN), 46.37% (RFC), and 36.75% (SVM). Based on prospective evaluation, the percentages of correctly classified cases across the 5 tasks were 60.11% (hard parameter sharing), 58.13% (cross-stitch), 51.30% (single-task CNN), 42.07% (RFC), and 35.16% (SVM). Moreover, hard parameter sharing MTCNNs outperformed the other models in computational efficiency by using about the same number of trainable parameters as a single-task CNN.ConclusionsThe hard parameter sharing MTCNN offers superior classification accuracy for automated coding support of pathology documents across a wide range of cancers and multiple information extraction tasks while maintaining similar training and inference time as those of a single task-specific model.
| S-EPMC7489089 | biostudies-literature
Project description:ImportanceCardiovascular disease is the leading cause of death in the United States. To improve cardiovascular outcomes, primary care must have valid methods of assessing performance on cardiovascular clinical quality measures, including aspirin use (aspirin measure), blood pressure control (BP measure), and smoking cessation counseling and intervention (smoking measure).ObjectiveTo compare observed performance scores measured using 2 imperfect reference standard data sources (medical record abstraction [MRA] and electronic health record [EHR]-generated reports) with misclassification-adjusted performance scores obtained using bayesian latent class analysis.Design, setting, and participantsThis cross-sectional study used a subset of the 2016 aspirin, BP, and smoking performance data from the Healthy Hearts for Oklahoma Project. Each clinical quality measure was calculated for a subset of a practice's patient population who can benefit from recommended care (ie, the eligible population). A random sample of 380 eligible patients were included for the aspirin measure; 126, for the BP measure; and 115, for the smoking measure. Data were collected from 21 primary care practices belonging to a single large health care system from January 1 to December 31, 2018, and analyzed from February 21 to April 17, 2019.Main outcomes and measuresThe main outcomes include performance scores for the aspirin, BP, and smoking measures using imperfect MRA and EHRs and estimated through bayesian latent class models.ResultsA total of 621 eligible patients were included in the analysis. Based on MRA and EHR data, observed aspirin performance scores were 76.0% (95% bayesian credible interval [BCI], 71.5%-80.1%) and 74.9% (95% BCI, 70.4%-79.1%), respectively; observed BP performance scores, 80.6% (95% BCI, 73.2%-86.9%) and 75.1% (95% BCI, 67.2%-82.1%), respectively; and observed smoking performance scores, 85.7% (95% BCI, 78.6%-91.2%) and 75.4% (95% BCI, 67.0%-82.6%), respectively. Misclassification-adjusted estimates were 74.9% (95% BCI, 70.5%-79.1%) for the aspirin performance score, 75.0% (95% BCI, 66.6%-82.5%) for the BP performance score, and 83.0% (95% BCI, 74.4%-89.8%) for the smoking performance score.Conclusions and relevanceEnsuring valid performance measurement is critical for value-based payment models and quality improvement activities in primary care. This study found that extracting information for the same individuals using different data sources generated different performance score estimates. Further research is required to identify the sources of these differences.
| S-EPMC7388024 | biostudies-literature
Project description:BackgroundRisk scores can be useful in clinical risk stratification and accurate allocations of medical resources, helping health providers improve patient care. Point-based scores are more understandable and explainable than other complex models and are now widely used in clinical decision making. However, the development of the risk scoring model is nontrivial and has not yet been systematically presented, with few studies investigating methods of clinical score generation using electronic health records.ObjectiveThis study aims to propose AutoScore, a machine learning-based automatic clinical score generator consisting of 6 modules for developing interpretable point-based scores. Future users can employ the AutoScore framework to create clinical scores effortlessly in various clinical applications.MethodsWe proposed the AutoScore framework comprising 6 modules that included variable ranking, variable transformation, score derivation, model selection, score fine-tuning, and model evaluation. To demonstrate the performance of AutoScore, we used data from the Beth Israel Deaconess Medical Center to build a scoring model for mortality prediction and then compared the data with other baseline models using the receiver operating characteristic analysis. A software package in R 3.5.3 (R Foundation) was also developed to demonstrate the implementation of AutoScore.ResultsImplemented on the data set with 44,918 individual admission episodes of intensive care, the AutoScore-created scoring models performed comparably well as other standard methods (ie, logistic regression, stepwise regression, least absolute shrinkage and selection operator, and random forest) in terms of predictive accuracy and model calibration but required fewer predictors and presented high interpretability and accessibility. The nine-variable, AutoScore-created, point-based scoring model achieved an area under the curve (AUC) of 0.780 (95% CI 0.764-0.798), whereas the model of logistic regression with 24 variables had an AUC of 0.778 (95% CI 0.760-0.795). Moreover, the AutoScore framework also drives the clinical research continuum and automation with its integration of all necessary modules.ConclusionsWe developed an easy-to-use, machine learning-based automatic clinical score generator, AutoScore; systematically presented its structure; and demonstrated its superiority (predictive performance and interpretability) over other conventional methods using a benchmark database. AutoScore will emerge as a potential scoring tool in various medical applications.
| S-EPMC7641783 | biostudies-literature