Practical use case of natural language processing for observational clinical research data retrieval from electronic health records: AssistMED project.

INTRODUCTION: Electronic health records (EHRs) contain data valuable for clinical research. However, they are in textual format and require manual encoding to databases, which is a lengthy and costly process. Natural language processing (NLP) is a computational technique that allows for text analysis. OBJECTIVES: Our study aimed to demonstrate a practical use case of NLP for a large retrospective study cohort characterization and comparison with human retrieval. PATIENTS AND METHODS: Anonymized discharge documentation of 10 314 patients from a cardiology tertiary care department was analyzed for inclusion in the CRAFT registry (Multicenter Experience in Atrial Fibrillation Patients Treated with Oral Anticoagulants; NCT02987062). Extensive clinical characteristics regarding concomitant diseases, medications, daily drug dosages, and echocardiography were collected manually and through NLP. RESULTS: There were 3030 and 3029 patients identified by human and NLP­based approaches, respectively, reflecting 99.93% accuracy of NLP in detecting AF. Comprehensive baseline patient characteristics by NLP was faster than human analysis (3 h and 15 min vs 71 h and 12 min). The calculated CHA2DS2VASc and HAS­BLED scores based on both methods did not differ (human vs NLP; median [interquartile range], 3 [2-5] vs 3 [2-5]; P = 0.74 and 1 [1-2] vs 1 [1-2]; P = 0.63, respectively). For most data, an almost perfect agreement between NLP- and human-retrieved characteristics was found; daily dosage identification was the least accurate NLP feature. Similar conclusions on cohort characteristics would be made; however, daily dosage detection for some drug groups would require additional human validation in the NLP­based cohort. CONCLUSIONS: NLP utilization in EHRs may accelerate data acquisition and provide accurate information for retrospective studies.

AssistMED project: Transforming cardiology cohort characterisation from electronic health records through natural language processing - Algorithm design, preliminary results, and field prospects.

INTRODUCTION: Electronic health records (EHR) are of great value for clinical research. However, EHR consists primarily of unstructured text which must be analysed by a human and coded into a database before data analysis- a time-consuming and costly process limiting research efficiency. Natural language processing (NLP) can facilitate data retrieval from unstructured text. During AssistMED project, we developed a practical, NLP tool that automatically provides comprehensive clinical characteristics of patients from EHR, that is tailored to clinical researchers needs. MATERIAL AND METHODS: AssistMED retrieves patient characteristics regarding clinical conditions, medications with dosage, and echocardiographic parameters with clinically oriented data structure and provides researcher-friendly database output. We validate the algorithm performance against manual data retrieval and provide critical quantitative and qualitative analysis. RESULTS: AssistMED analysed the presence of 56 clinical conditions, medications from 16 drug groups with dosage and 15 numeric echocardiographic parameters in a sample of 400 patients hospitalized in the cardiology unit. No statistically significant differences between algorithm and human retrieval were noted. Qualitative analysis revealed that disagreements with manual annotation were primarily accounted to random algorithm errors, erroneous human annotation and lack of advanced context awareness of our tool. CONCLUSIONS: Current NLP approaches are feasible to acquire accurate and detailed patient characteristics tailored to clinical researchers' needs from EHR. We present an in-depth description of an algorithm development and validation process, discuss obstacles and pinpoint potential solutions, including opportunities arising with recent advancements in the field of NLP, such as large language models.

Remotely supervised spirometry versus laboratory-based spirometry during the COVID-19 pandemic: a retrospective analysis.

BACKGROUND: The COVID-19 pandemic has constrained access to spirometry, and the inherent risk of infectious transmission during aerosol-generating procedures has necessitated the rapid development of Remotely Supervised Spirometry (RSS). This innovative approach enables patients to perform spirometry tests at home, using a mobile connected spirometer, all under the real-time supervision of a technician through an online audio or video call. METHODS: In this retrospective study, we examined the quality of RSS in comparison to conventional Laboratory-based Spirometry (LS), using the same device and technician. Our sample included 242 patients, with 129 undergoing RSS and 113 participating in LS. The RSS group comprised 51 females (39.5%) with a median age of 37 years (range: 13-76 years). The LS group included 63 females (55.8%) with a median age of 36 years (range: 12-80 years). RESULTS: When comparing the RSS group to the LS group, the percentage of accurate Forced Expiratory Volume in one second (FEV1) measurements was 78% (n = 101) vs. 86% (n = 97), p = 0.177; for Forced Vital Capacity (FVC) it was 77% (n = 99) vs. 82% (n = 93), p = 0.365; and for both FEV1 and FVC, it was 75% (n = 97) vs. 81% (n = 92), p = 0.312, respectively. CONCLUSIONS: Our findings demonstrate no significant difference in the quality of spirometry testing between RSS and LS, a result that held true across all age groups, including patients aged over 65 years. The principal advantages of remote spirometry include improved access to pulmonary function tests, reduced infectious risk to curtail disease spread, and enhanced convenience for patients.

Deep learning algorithm for visual quality assessment of the spirograms.

Objective. The quality of spirometry manoeuvres is crucial for correctly interpreting the values of spirometry parameters. A fundamental guideline for proper quality assessment is the American Thoracic Society and European Respiratory Society (ATS/ERS) Standards for spirometry, updated in 2019, which describe several start-of-test and end-of-test criteria which can be assessed automatically. However, the spirometry standards also require a visual evaluation of the spirometry curve to determine the spirograms' acceptability or usability. In this study, we present an automatic algorithm based on a convolutional neural network (CNN) for quality assessment of the spirometry curves as an alternative to manual verification performed by specialists.Approach. The algorithm for automatic assessment of spirometry measurements was created using a set of randomly selected 1998 spirograms which met all quantitative criteria defined by ATS/ERS Standards. Each spirogram was annotated as 'confirm' (remaining acceptable or usable status) or 'reject' (change the status to unacceptable) by four pulmonologists, separately for FEV1 and FVC parameters. The database was split into a training (80%) and test set (20%) for developing the CNN classification algorithm. The algorithm was optimised using a cross-validation method.Main results. The accuracy, sensitivity and specificity obtained for the algorithm were 92.6%, 93.1% and 90.0% for FEV1 and 94.1%, 95.6% and 88.3% for FVC, respectively.Significance.The algorithm provides an opportunity to significantly improve the quality of spirometry tests, especially during unsupervised spirometry. It can also serve as an additional tool in clinical trials to quickly assess the quality of a large group of tests.

Photoplethysmography wave morphology in patients with atrial fibrillation.

Objective.Most current algorithms for detecting atrial fibrillation (AF) rely on heart rate variability (HRV), and only a few studies analyse the variability of photopletysmography (PPG) waveform. This study aimed to compare morphological features of the PPG curve in patients with AF to those presenting a normal sinus rhythm (NSR) and evaluate their usefulness in AF detection.Approach.10 min PPG signals were obtained from patients with persistent/paroxysmal AF and NSR. Nine morphological parameters (1/ΔT), Pulse Width [PW], augmentation index [AI], b/a, e/a, [b-e]/a, crest time [CT], inflection point area [IPA], Area and five HRV parameters (heart rate [HR], Shannon entropy [ShE], root mean square of the successive differences [RMSSD], number of pairs of consecutive systolic peaks [R-R] that differ by more than 50 ms [NN50], standard deviation of theR-Rintervals [SDNN]) were calculated.Main results.Eighty subjects, including 33 with AF and 47 with NSR were recruited. In univariate analysis five morphological features (1/ΔT,p< 0.001; b/a,p< 0.001; [b-e]/a,p< 0.001; CT,p= 0.011 and Area,p< 0.001) and all HRV parameters (p= 0.01 for HR andp< 0.001 for others) were significantly different between the study groups. In the stepwise multivariate model (Area under the curve [AUC] = 0.988 [0.974-1.000]), three morphological parameters (PW,p< 0.001; e/a,p= 0.011; (b-e)/a,p< 0.001) and three of HRV parameters (ShE,p= 0.01; NN50,p< 0.001, HR,p= 0.01) were significant.Significance.There are significant differences between AF and NSR, PPG waveform, which are useful in AF detection algorithm. Moreover adding those features to HRV-based algorithms may improve their specificity and sensitivity.

Flecainide in clinical practice.

Flecainide, similar to encainide and propafenone, is IC class antiarrhythmic, inhibiting Nav1.5 sodium channels in heart muscle cells and modulates cardiac conduction. Despite its over 40-year presence in clinical practice, strong evidence and well-known safety profile, flecainide distribution in Europe has not always been equal. In Poland, the drug has been available in pharmacies only since October this year, and previously it had to be imported on request. Flecainide can be used successfully in both the acute and chronic treatment of cardiac arrhythmias. The main indication for flecainide is the treatment of paroxysmal supraventricular tachycardias, including atrial fibrillation, atrioventricular nodal re-entrant tachycardia, atrioventricular re-entrant tachycardia and ventricular arrhythmias in patients without structural heart disease. Beyond that, it may be used in some supraventricular tachycardia in children and for sustained fetal tachycardia. Many studies indicate its efficacy comparable to or better than previously used drugs such as propafenone and amiodarone, depending on the indication. This review aims to highlight the most important clinical uses of flecainide in the light of the latest scientific evidence and to provide an overview of the practical aspects of treatment, including indications, off-label use, contraindications, areas of use, monitoring of treatment and most common complications, taking into account special populations: children and pregnant women.

Administrative Data in Cardiovascular Research-A Comparison of Polish National Health Fund and CRAFT Registry Data.

(1) Background: Administrative data allows for time- and cost-efficient acquisition of large volumes of individual patient data invaluable for evaluation of the prevalence of diseases and clinical outcomes. The aim of the study was to evaluate the accuracy of data collected from the Polish National Health Fund (NHF), from a researcher's perspective, in regard to a cohort of atrial fibrillation patients. (2) Methods: NHF data regarding atrial fibrillation and common cardiovascular comorbidities was compared with the data collected manually from the individual patients' health records (IHR) collected in the retrospective CRAFT registry (NCT02987062). (3) Results: Data from the NHF underestimated the proportion of patients with AF (NHF = 83% vs. IHR = 100%) while overestimating the proportion of patients with other cardiovascular comorbidities in the cohort. Significantly higher CHA2DS2VASc (Median, [Q1-Q3]) (NHF: 1, [0-2]; vs. IHR: 1, [0-1]; p < 0.001) and HAS-BLED (Median, [Q1-Q3]) (NHF: 4, [2-6] vs. IHR: 3, [2-5]; p < 0.001) scores were calculated according to NHF in comparison to IHR data, respectively. (4) Conclusions: Clinical researchers should be aware that significant differences between IHR and billing data in cardiovascular research can be observed which should be acknowledged while drawing conclusions from administrative data-based cohorts. Natural Language Processing of IHR could further increase administrative data quality in the future.