Development and Evaluation of a Natural Language Processing System for Curating a Trans-Thoracic Echocardiogram (TTE) Database.

BACKGROUND: Although electronic health records (EHR) provide useful insights into disease patterns and patient treatment optimisation, their reliance on unstructured data presents a difficulty. Echocardiography reports, which provide extensive pathology information for cardiovascular patients, are particularly challenging to extract and analyse, because of their narrative structure. Although natural language processing (NLP) has been utilised successfully in a variety of medical fields, it is not commonly used in echocardiography analysis. OBJECTIVES: To develop an NLP-based approach for extracting and categorising data from echocardiography reports by accurately converting continuous (e.g., LVOT VTI, AV VTI and TR Vmax) and discrete (e.g., regurgitation severity) outcomes in a semi-structured narrative format into a structured and categorised format, allowing for future research or clinical use. METHODS: 135,062 Trans-Thoracic Echocardiogram (TTE) reports were derived from 146967 baseline echocardiogram reports and split into three cohorts: Training and Validation (n = 1075), Test Dataset (n = 98) and Application Dataset (n = 133,889). The NLP system was developed and was iteratively refined using medical expert knowledge. The system was used to curate a moderate-fidelity database from extractions of 133,889 reports. A hold-out validation set of 98 reports was blindly annotated and extracted by two clinicians for comparison with the NLP extraction. Agreement, discrimination, accuracy and calibration of outcome measure extractions were evaluated. RESULTS: Continuous outcomes including LVOT VTI, AV VTI and TR Vmax exhibited perfect inter-rater reliability using intra-class correlation scores (ICC = 1.00, p < 0.05) alongside high R2 values, demonstrating an ideal alignment between the NLP system and clinicians. A good level (ICC = 0.75-0.9, p < 0.05) of inter-rater reliability was observed for outcomes such as LVOT Diam, Lateral MAPSE, Peak E Velocity, Lateral E' Velocity, PV Vmax, Sinuses of Valsalva and Ascending Aorta diameters. Furthermore, the accuracy rate for discrete outcome measures was 91.38% in the confusion matrix analysis, indicating effective performance. CONCLUSIONS: The NLP-based technique yielded good results when it came to extracting and categorising data from echocardiography reports. The system demonstrated a high degree of agreement and concordance with clinician extractions. This study contributes to the effective use of semi-structured data by providing a useful tool for converting semi-structured text to a structured echo report that can be used for data management. Additional validation and implementation in healthcare settings can improve data availability and support research and clinical decision-making.

Maternal metabolomic profiling and congenital heart disease risk in offspring: A systematic review of observational studies.

Aetiological understanding and screening methods for congenital heart disease (CHD) are limited. Maternal metabolomic assessment offers the potential to identify risk factors and biomarkers. We performed a systematic review (PROSPERO CRD42022308452) investigating the association between fetal/childhood CHD and endogenous maternal metabolites. Ovid-MEDLINE, Ovid-EMBASE and Cochrane Library were searched between inception and 06/09/2022. Case control studies included analysing maternal blood or urine metabolites in pregnancy or postpartum where there was foetal/childhood CHD. Risk of bias assessment utilised the Scottish Intercollegiate Guidelines Network methodology checklist and narrative synthesis was performed. A total of 134 records were screened with eight eligible studies (n = 3242 pregnancies, n = 842 CHD-affected offspring). Five studies performed metabolomic analysis in pregnancy. Metabolites distinguishing case and control groups spanned lipid, glucose and amino-acid pathways, with the development of sensitive risk prediction models. No single metabolite consistently distinguished cases and controls across studies. Three studies performed targeted analysis postnatally with altered lipid and amino acid metabolites and raised homocysteine and markers of oxidative stress identified in cases. Included studies reported small sample sizes, analysing different biosamples at variable time points using differing techniques. At present, there is not enough evidence to confidently associate maternal metabolomic profiles with offspring CHD risk. However, several identified pathways warrant further investigation.

Maternal micronutrient deficiency and congenital heart disease risk: A systematic review of observational studies.

BACKGROUND: Congenital anomalies affect over 2% of pregnancies, with congenital heart disease (CHD) the most common. Understanding of causal factors is limited. Micronutrients are essential trace elements with key roles in growth and development. We aimed to investigate whether maternal micronutrient deficiencies increase the risk of fetal CHD through systematic review of published literature. METHOD: We performed a systematic review registered at PROSPERO as CRD42021276699. Ovid-MEDLINE, Ovid-EMBASE, and Cochrane Library were searched from their inception until September 7, 2021. Case control trials were included with a population of biological mothers of fetuses with and without CHD. The exposure was maternal micronutrient level measured in pregnancy or the postpartum period. Data extraction was performed by one author and checked by a second. Risk of bias assessment was performed according to the Scottish Intercollegiate Guidelines Network guidance. We performed a narrative synthesis for analysis. RESULTS: 726 articles were identified of which 8 met our inclusion criteria. Final analysis incorporated data from 2,427 pregnancies, 1,199 of which were complicated by fetal CHD assessing 8 maternal micronutrients: vitamin D, vitamin B12, folate, vitamin A, zinc, copper, selenium, and ferritin. Studies were heterogenous with limited sample sizes and differing methods and timing of maternal micronutrient sampling. Definitions of deficiency varied and differed from published literature. Published results were contradictory. CONCLUSION: There is not enough evidence to confidently conclude if maternal micronutrient deficiencies increase the risk of fetal CHD. Further large-scale prospective study is required to answer this question.

Surgical-PEARL protocol: a multicentre prospective cohort study exploring aetiology, management and outcomes for patients with congenital anomalies potentially requiring surgical intervention.

INTRODUCTION: Congenital anomalies affect over 2% of pregnancies. Surgical advances have reduced mortality and improved survival for patients with congenital anomalies potentially requiring surgical (CAPRS) intervention. However, our understanding of aetiology, diagnostic methods, optimal management, outcomes and prognostication is limited. Existing birth cohorts have low numbers of individual heterogenous CAPRS. The Surgical Paediatric congEnital Anomalies Registry with Long term follow-up (Surgical-PEARL) study aims to establish a multicentre prospective fetal, child and biological parent cohort of CAPRS. METHODS AND ANALYSIS: From 2022 to 2027, Surgical-PEARL aims to recruit 2500 patients with CAPRS alongside their biological mothers and fathers from up to 15 UK centres. Recruitment will be antenatal or postnatal dependent on diagnosis timing and presentation to a recruitment site. Routine clinical data including antenatal scans and records, neonatal intensive care unit (NICU) records, diagnostic and surgical data and hospital episode statistics will be collected. A detailed biobank of samples will include: parents' blood and urine samples; amniotic fluid if available; children's blood and urine samples on admission to NICU, perioperatively or if the child has care withdrawn or is transferred for extracorporeal membrane oxygenation; stool samples; and surplus surgical tissue. Parents will complete questionnaires including sociodemographic and health data. Follow-up outcome and questionnaire data will be collected for 5 years. Once established we will explore the potential of comparing findings in Surgical-PEARL to general population cohorts born in the same years and centres. ETHICS AND DISSEMINATION: Ethical and health research authority approvals have been granted (IRAS Project ID: 302251; REC reference number 22/SS/0004). Surgical-PEARL is adopted onto the National Institute for Health Research Clinical Research Network portfolio. Findings will be disseminated widely through peer-reviewed publication, conference presentations and through patient organisations and newsletters. TRIAL REGISTRATION NUMBER: ISRCTN12557586.