The goal of this observational study is to learn whether combining stress echocardiography (stress echo) results with routine clinical information can better predict important heart outcomes in adults (18+) with chest pain who were assessed for suspected coronary artery disease.

The main questions it aims to answer are:

Can an artificial intelligence / machine learning model using stress echo findings plus clinical factors (such as blood pressure, diabetes, smoking, other health conditions, medications, and body measurements) predict major heart-related events (such as heart attack, stroke, death related to heart disease, or the need for coronary procedures) more accurately than stress echo results alone?

Can the model help identify which patients are most likely to benefit from further invasive assessment and possible coronary revascularisation (for example, a stent or bypass surgery)?

Which combination of stress echo measurements and clinical factors contributes most to risk prediction?

Participants will:

Not be asked to attend extra visits or have additional tests for this study.

Have their existing stress echo reports and routinely collected hospital record data analysed (approximately 3,000 people who previously had dobutamine stress echo at Milton Keynes University Hospital).

In some cases, if outcomes are not fully available from hospital records, the research team may check additional sources (such as GP records, or contacting the patient if appropriate) to confirm whether a major heart-related event occurred.

This is a single-centre, retrospective observational study using an existing dataset of pharmacological (dobutamine) stress echocardiography (SE) reports generated within Milton Keynes University Hospital over approximately 15 years, starting from 2002. The SE dataset comprises reports/letters produced by a single, experienced clinician, which reduces inter-observer variability and supports consistent interpretation across the cohort.

Data sources and cohort construction

SE reports (in document format) will be converted into a structured research database. A computer science team will develop a generalisable approach to extract structured variables from the clinical SE reports, building on prior proof-of-concept work demonstrating feasibility of converting these reports into a database.

The dataset includes clinical variables (e.g., cardiovascular risk factors, comorbidities, prescribed medications, and anthropometrics) alongside SE-derived measures (including ischaemia detection and wall motion scoring at rest and peak stress).

Stress echocardiography technique (context for imaging-derived variables)

The study dataset reflects contemporary dobutamine SE practice at MKUH, with contrast-enhanced imaging used in the majority of cases (SonoVue contrast with rota pump infusion equipment). Studies were performed predominantly on Philips echocardiography systems, with image acquisition across standard stages (resting, intermediate, peak stress, and recovery) and standard views (apical 4-, 2-, and 3-chamber; parasternal long- and short-axis). Reporting used dedicated platforms enabling stage-by-stage comparison.

Outcome ascertainment and linkage

Following database completion, a research nurse will query the hospital Electronic Data Management system to ascertain major adverse cardiovascular events (MACE) for the cohort. Where outcomes cannot be confirmed from hospital systems (e.g., patients no longer served by the hospital), missing outcome information will be explored via primary care physician contact and/or patient contact as appropriate.

Data processing, quality checks, and handling missingness

Extracted data will undergo cleaning prior to analysis. Natural Language Processing (NLP) and feature engineering approaches will be used to transform extracted information into model-ready features. As part of preprocessing, data fields will be checked for completeness and consistency before modelling. Missing outcome data will be addressed through the external outcome checks described above.

Statistical / machine learning approach and internal validation

After preprocessing, subset feature selection methods will be applied to identify the most informative predictors for risk classification. Supervised learning will be used to discriminate between lower-risk cases and cases requiring further investigation, with additional modelling approaches (including regression techniques) planned to support quantification of disease stage in abnormal cases. Overfitting will be mitigated through use of techniques robust to overfitting (e.g., ensemble methods) and internal validation using k-fold cross-validation (five folds), ensuring separation of training and validation data.

Sample size and additional analyses

The study will utilise the available full dataset (approximately 3,000 patients) to maximise model development and internal validation. A cost analysis is also planned using the available data.

Inclusion Criteria:

• Age 18 years or older at the time of the index stress echocardiography.
• Referred for pharmacological (dobutamine) stress echocardiography at Milton Keynes University Hospital for assessment of suspected coronary artery disease / chest pain.
• Stress echocardiography report available in the hospital dataset for data extraction and conversion into a structured database.

Exclusion Criteria:

• Age under 18 years at the time of the index stress echocardiography.
• No available/usable stress echocardiography report for extraction into the study database.
• Unable to link the record to follow-up outcome information using routine hospital systems (with attempted supplementary checks where needed).
• Patients who have registered a National Data Opt-out and are therefore not eligible for use of their confidential patient information for research/secondary purposes in this study.

Single-centre retrospective cohort drawn from Milton Keynes University Hospital records, consisting of approximately 3,000 adults who underwent clinically indicated pharmacological (dobutamine) stress echocardiography for assessment of chest pain / suspected coronary artery disease (dataset dating back to 2002, covering ~15 years of reports). Stress echocardiography report variables are extracted and linked to routine clinical outcome data for analysis.

De-identified, disclosure-controlled individual participant data will be made available to external researchers for accredited research purposes / research in the public interest. Data shared will be limited to the minimum necessary variables to meet the approved research purpose and will be processed to reduce re-identification risk (e.g., de-identification and disclosure control, including suppression/controls where needed). Data will only be released once the approvals and agreements described in the access criteria are in place.

Requests may be submitted once the research proposal is ready for review. A decision is expected within ~21 days of receipt of a complete application (including all required supporting documents and agreements). Approved data will be transferred after sign-off and completion of the Third Party Agreement, and availability will be case-by-case subject to the database being maintained and the governance safeguards remaining appropriate.

Access is not open access. Researchers must apply via the Trust's R&D contact route with full details of the proposed research, justification, and the specific data required. The applicant must be the Principal Investigator for the proposed project and provide evidence of suitability (e.g., CV and GCP certificate where relevant) and enter into the Trust's third-party data access agreement. Requests are reviewed through the Trust governance process, including confirmation that appropriate peer review, patient/public involvement and ethics/regulatory approvals are in place where required. Final information governance sign-off is required before release. Transfers are completed by the database/data custodian and recorded (data transferred, format, and date).

NOTE: Some items (incl. outcome data received from NHS England) may need an MKUH-NHSE contract amendment to name the requester and may incur charges. Contact MKUH R&D: research@mkuh.nhs.uk for further information.