New platform unlocks clinical intelligence from routine hospital data

Developed at Great Ormond Street Hospital (GOSH), the PICTURE platform provides clinicians, managers and researchers with a scalable way to define patient cohorts, extract relevant EHR data and generate analyses that can support clinical decision-making, research and operational planning. The study argues that the approach could reshape how healthcare providers think about and use their data.

The research team, led by chief research information officer Professor Neil Sebire and colleagues at the GOSH DRIVE Unit, created PICTURE to address a long-standing challenge: although hospitals hold vast quantities of clinical data, these are rarely accessible in a way that allows staff to answer practical questions about patient care or service delivery.

‘Electronic health records have transformed patient care, but turning routine data into meaningful insight remains difficult for most clinicians,’ the authors write. ‘Our platform demonstrates a generalisable approach to making that insight available on demand.’

PICTURE can generate interactive visualisations or static PDF reports and is designed to integrate machine-learning or third-party analytics tools in future.

To demonstrate practical value, the authors used PICTURE to answer diverse clinical and operational questions using paediatric nephrology and cardiology datasets. Examples included:

• comparing post-transplant kidney function in an individual patient against historic cohorts
• identifying the most common medications used in dialysis patients
• exploring risk factors for pleural effusion after cardiac surgery
• analysing time to postoperative echocardiography after ventricular septal defect repair.

These use cases show how routine data can quickly reveal patterns, variations and operational behaviours that would otherwise remain hidden.

The authors also highlight challenges, including the need for high-quality data, interoperability between EHR systems, and careful interpretation of results – particularly given the distinction between statistical associations and clinical causality.

The study concludes that platforms like PICTURE could help healthcare providers make better use of the data they already collect every day.

‘With further development, this approach has the potential to democratise, promote and simplify access to aggregated EHR data to improve patient care,’ say the authors.

The research was carried out as part of an ongoing NHS-Industry partnership between GOSH and Roche Products Ltd and was supported by Great Ormond Street Hospital Children’s Charity and the National Institute for Health and Care Research GOSH Biomedical Research Centre.

Interested in this? You may like…

RCP View on digital and AI

Seven in ten physicians tell us the NHS is not digitally fit to deploy artificial intelligence (AI) safely or effectively. Our new RCP View on digital and AI sets out what clinicians need from national and system leaders to unlock the benefits of AI while protecting patients. Many NHS staff are already using AI in their daily practice, often without clear guidance or assurance. The message is clear: we need decisive leadership to reduce variation, manage risk and maintain confidence in the system. Our report calls for:

• urgent action to fix core digital infrastructure and interoperability
• AI solutions that solve real clinical problems, not abstract use cases
• a clear, NHS-approved framework for AI tools, with joined-up regulation to safeguard patients and clinicians.