Structural MRI of the brain is routinely performed on patients referred to memory clinics; however, resulting radiology reports, including volumetric assessments, are conventionally stored as unstructured free text. We sought to use natural language processing (NLP) to extract text relating to intracranial volumetric assessment from brain MRI text reports to enhance routine data availability for research purposes.

Electronic records from a large mental healthcare provider serving a geographic catchment of 1.3 million residents in four boroughs of south London, UK.

A corpus of 4007 de-identified brain MRI reports from patients referred to memory assessment services. An NLP algorithm was developed, using a span categorisation approach, to extract six binary (presence/absence) categories from the text reports: (i) global volume loss, (ii) hippocampal/medial temporal lobe volume loss and (iii) other lobar/regional volume loss. Distributions of these categories were evaluated.

The overall F1 score for the six categories was 0.89 (precision 0.92, recall 0.86), with the following precision/recall for each category: presence of global volume loss 0.95/0.95, absence of global volume loss 0.94/0.77, presence of regional volume loss 0.80/0.58, absence of regional volume loss 0.91/0.93, presence of hippocampal volume loss 0.90/0.88, and absence of hippocampal volume loss 0.94/0.92.

These results support the feasibility and accuracy of using NLP techniques to extract volumetric assessments from radiology reports, and the potential for automated generation of novel meta-data from dementia assessments in electronic health records.

Persistent musculoskeletal pain is a leading cause of disability and need for rehabilitation globally. Many people with the condition attend pain management programmes (PMPs) for rehabilitation and support with self-management. Physical activity (PA) is an essential self-management strategy facilitated on PMPs as it benefits symptoms, general health and well-being. PA needs to be maintained in the long term to continue to be beneficial. However, while many patients increase their PA during or immediately after a PMP, they commonly find it difficult to maintain it in the long term. This study aims to address this problem by developing an intervention to support PA maintenance after a PMP.

This mixed-methods study will be guided by the Medical Research Council guidelines for developing complex interventions and the Behaviour Change Wheel intervention development framework. Participants will be recruited from multiple UK National Health Service PMPs. Participants will include patients with persistent musculoskeletal pain who have completed PMPs, their PA partners (people who support them with PA) and healthcare professionals who facilitate PA on PMPs. The study will be conducted in three phases. In phase 1, qualitative interviews will explore the experiences, barriers and facilitators of PA maintenance after a PMP and potential characteristics for a PA maintenance intervention from patient, PA partner and healthcare professional perspectives. Phase 2 will consist of a prospective longitudinal pilot study to identify factors associated with PA maintenance after a PMP. Phase 3 will involve developing a logic model and co-designing the intervention with patient, PA partner and healthcare professional stakeholder groups.

The project received research ethics committee (REC) and Health Research Authority approval on 4 June 2024 (REC: North West—Liverpool Central, REC reference: 24/NW/0174, IRAS Project ID: 340674). Findings will be disseminated by peer-reviewed publications, conference presentations, social media and lay summaries for patients and the public.