Emma M. Davidson
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Emma M. Davidson has a total of 71 co-authors. Here are the top co-authors by number of shared publications:
- Erik Beune
Department of Public & Occupational Health, Amsterdam Public Health Institute, Amsterdam UMC, Location AMC, University of Amsterdam, Amsterdam, the Netherlands
3 shared publications - Esperanza Diaz
Pandemic Center, Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway
3 shared publications - William Whiteley
University of Edinburgh
3 shared publications - Anne Karen Jenum
Department of General Practice, Institute of Health and Society, University of Oslo, Oslo, Norway
2 shared publications Arlene Casey
Advanced Care Research Centre, Usher Institute, University of Edinburgh
2 shared publications- Aziz Sheikh
Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK
2 shared publications
Latest Publications
- Open Access
Clinically reported covert cerebrovascular disease and risk of neurological disease: a whole-population cohort of 367,988 people using natural language processing
Importance Understanding the relevance of covert cerebrovascular disease (CCD) for later health will allow clinicians to more effectively monitor and target interventions. Objective To examine the association between clinically reported CCD, measured using natural language processing (NLP), and subsequent disease risk. Design, Setting and Participants We conducted a retrospective e-cohort study using linked health record data. From all people with clinical brain imaging in Scotland from 2010 to 2018, we selected people with no prior hospitalisation for neurological disease. The data were analysed from March 2024 to June 2025. Exposure Four phenotypes were identified with NLP of imaging reports: white matter hypoattenuation or hyperintensities (WMH), lacunes, cortical infarcts and cerebral atrophy. Main outcomes and measures Hazard ratios (aHR) for stroke, dementia, and Parkinson’s disease (conditions previously associated with CCD), epilepsy (a brain-based control condition) and colorectal cancer (a non-brain control condition), adjusted for age, sex, deprivation, region, scan modality, and pre-scan healthcare, were calculated for each phenotype. Results From 395,273 people with brain imaging and no history of neurological disease, 145,978 (37%) had ≥1 phenotype. For each phenotype, the aHR of any stroke was: WMH 1.4 (95%CI: 1.3–1.4), lacunes 1.6 (1.5–1.6), cortical infarct 1.7 (1.6–1.8), and cerebral atrophy 1.1 (1.0–1.1). The aHR of any dementia was: WMH, 1.3 (1.3–1.3), lacunes, 1.0 (0.9–1.0), cortical infarct 1.1 (1.0–1.1) and cerebral atrophy 1.7 (1.7–1.7). The aHR of Parkinson’s disease was, in people with a report of: WMH 1.1 (1.0–1.2), lacunes 1.1 (0.9–1.2), cortical infarct 0.7 (0.6–0.9) and cerebral atrophy 1.4 (1.3–1.5). The aHRs between CCD phenotypes and epilepsy and colorectal cancer overlapped the null. Conclusions and Relevance NLP identified CCD and atrophy phenotypes from routine clinical image reports, and these had important associations with future stroke, dementia and Parkinson’s disease. Prevention of neurological disease in people with CCD should be a priority for healthcare providers and policymakers. Question Are measures of Covert Cerebrovascular Disease (CCD) associated with the risk of subsequent disease (stroke, dementia, Parkinson’s disease, epilepsy, and colorectal cancer)? Findings This study used a validated NLP algorithm to identify CCD (white matter hypoattenuation/hyperintensities, lacunes, cortical infarcts) and cerebral atrophy from both MRI and computed tomography (CT) imaging reports generated during routine healthcare in >395K people in Scotland. In adjusted models, we demonstrate higher risk of dementia (particularly Alzheimer’s disease) in people with atrophy, and higher risk of stroke in people with cortical infarcts. However, associations with an age-associated control outcome (colorectal cancer) were neutral, supporting a causal relationship. It also highlights differential associations between cerebral atrophy and dementia and cortical infarcts and stroke risk. Meaning CCD or atrophy on brain imaging reports in routine clinical practice is associated with a higher risk of stroke or dementia. Evidence is needed to support treatment strategies to reduce this risk. NLP can identify these important, otherwise uncoded, disease phenotypes, allowing research at scale into imaging-based biomarkers of dementia and stroke.
Open AccessClinically reported covert cerebrovascular disease and risk of neurological disease: a whole-population cohort of 367,988 people using natural language processing
Importance Understanding the relevance of covert cerebrovascular disease (CCD) for later health will allow clinicians to more effectively monitor and target interventions. Objective To examine the association between clinically reported CCD, measured using natural language processing (NLP), and subsequent disease risk. Design, Setting and Participants We conducted a retrospective e-cohort study using linked health record data. From all people with clinical brain imaging in Scotland from 2010 to 2018, we selected people with no prior hospitalisation for neurological disease. The data were analysed from March 2024 to June 2025. Exposure Four phenotypes were identified with NLP of imaging reports: white matter hypoattenuation or hyperintensities (WMH), lacunes, cortical infarcts and cerebral atrophy. Main outcomes and measures Hazard ratios (aHR) for stroke, dementia, and Parkinson’s disease (conditions previously associated with CCD), epilepsy (a brain-based control condition) and colorectal cancer (a non-brain control condition), adjusted for age, sex, deprivation, region, scan modality, and pre-scan healthcare, were calculated for each phenotype. Results From 395,273 people with brain imaging and no history of neurological disease, 145,978 (37%) had ≥1 phenotype. For each phenotype, the aHR of any stroke was: WMH 1.4 (95%CI: 1.3–1.4), lacunes 1.6 (1.5–1.6), cortical infarct 1.7 (1.6–1.8), and cerebral atrophy 1.1 (1.0–1.1). The aHR of any dementia was: WMH, 1.3 (1.3–1.3), lacunes, 1.0 (0.9–1.0), cortical infarct 1.1 (1.0–1.1) and cerebral atrophy 1.7 (1.7–1.7). The aHR of Parkinson’s disease was, in people with a report of: WMH 1.1 (1.0–1.2), lacunes 1.1 (0.9–1.2), cortical infarct 0.7 (0.6–0.9) and cerebral atrophy 1.4 (1.3–1.5). The aHRs between CCD phenotypes and epilepsy and colorectal cancer overlapped the null. Conclusions and Relevance NLP identified CCD and atrophy phenotypes from routine clinical image reports, and these had important associations with future stroke, dementia and Parkinson’s disease. Prevention of neurological disease in people with CCD should be a priority for healthcare providers and policymakers. Question Are measures of Covert Cerebrovascular Disease (CCD) associated with the risk of subsequent disease (stroke, dementia, Parkinson’s disease, epilepsy, and colorectal cancer)? Findings This study used a validated NLP algorithm to identify CCD (white matter hypoattenuation/hyperintensities, lacunes, cortical infarcts) and cerebral atrophy from both MRI and computed tomography (CT) imaging reports generated during routine healthcare in >395K people in Scotland. In adjusted models, we demonstrate higher risk of dementia (particularly Alzheimer’s disease) in people with atrophy, and higher risk of stroke in people with cortical infarcts. However, associations with an age-associated control outcome (colorectal cancer) were neutral, supporting a causal relationship. It also highlights differential associations between cerebral atrophy and dementia and cortical infarcts and stroke risk. Meaning CCD or atrophy on brain imaging reports in routine clinical practice is associated with a higher risk of stroke or dementia. Evidence is needed to support treatment strategies to reduce this risk. NLP can identify these important, otherwise uncoded, disease phenotypes, allowing research at scale into imaging-based biomarkers of dementia and stroke.
- Open Access
- 15 September 2023
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 7 (1) , e33810
A Culturally Adapted Diet and Physical Activity Text Message Intervention to Prevent Type 2 Diabetes Mellitus for Women of Pakistani Origin Living in Scotland: Formative Study
Background: Individuals of South Asian origin are at an increased risk of developing type 2 diabetes mellitus (T2DM) compared with other ethnic minority groups. Therefore, there is a need to develop interventions to address, and reduce, this heightened risk. Objective: We undertook formative work to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. Methods: We used a stepwise approach that was informed by the Six Steps in Quality Intervention Development framework, which consisted of gathering evidence through literature review and focus groups (step 1), developing a program theory for the intervention (step 2), and finally developing the content of the text messages and an accompanying delivery plan (step 3). Results: In step 1, we reviewed 12 articles and identified 3 key themes describing factors impacting on diet and physical activity in the context of T2DM prevention: knowledge on ways to prevent T2DM through diet and physical activity; cultural, social, and gender norms; and perceived level of control and sense of inevitability over developing T2DM. The key themes that emerged from the 3 focus groups with a total of 25 women were the need for interventions to provide “friendly encouragement,” “companionship,” and a “focus on the individual” and also for the text messages to “set achievable goals” and include “information on cooking healthy meals.” We combined the findings of the focus groups and literature review to create 13 guiding principles for culturally adapting the text messages. In step 2, we developed a program theory, which specified the main determinants of change that our text messages should aim to enhance: knowledge and skills, sense of control, goal setting and planning behavior, peer support, and norms and beliefs guiding behavior. In step 3, we used both the intervention program theory and guiding principles to develop a set of 73 text messages aimed at supporting a healthy diet and 65 text messages supporting increasing physical activity. Conclusions: We present a theory-based approach to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. This study outlines an approach that may also be applicable to the development of interventions for other ethnic minority populations in diverse settings. There is now a need to build on this formative work and undertake a feasibility trial of a text message–based diet and physical activity intervention to prevent T2DM for women of Pakistani origin living in Scotland.
- 15 September 2023
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 7 (1) , e33810
Open AccessA Culturally Adapted Diet and Physical Activity Text Message Intervention to Prevent Type 2 Diabetes Mellitus for Women of Pakistani Origin Living in Scotland: Formative Study
Background: Individuals of South Asian origin are at an increased risk of developing type 2 diabetes mellitus (T2DM) compared with other ethnic minority groups. Therefore, there is a need to develop interventions to address, and reduce, this heightened risk. Objective: We undertook formative work to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. Methods: We used a stepwise approach that was informed by the Six Steps in Quality Intervention Development framework, which consisted of gathering evidence through literature review and focus groups (step 1), developing a program theory for the intervention (step 2), and finally developing the content of the text messages and an accompanying delivery plan (step 3). Results: In step 1, we reviewed 12 articles and identified 3 key themes describing factors impacting on diet and physical activity in the context of T2DM prevention: knowledge on ways to prevent T2DM through diet and physical activity; cultural, social, and gender norms; and perceived level of control and sense of inevitability over developing T2DM. The key themes that emerged from the 3 focus groups with a total of 25 women were the need for interventions to provide “friendly encouragement,” “companionship,” and a “focus on the individual” and also for the text messages to “set achievable goals” and include “information on cooking healthy meals.” We combined the findings of the focus groups and literature review to create 13 guiding principles for culturally adapting the text messages. In step 2, we developed a program theory, which specified the main determinants of change that our text messages should aim to enhance: knowledge and skills, sense of control, goal setting and planning behavior, peer support, and norms and beliefs guiding behavior. In step 3, we used both the intervention program theory and guiding principles to develop a set of 73 text messages aimed at supporting a healthy diet and 65 text messages supporting increasing physical activity. Conclusions: We present a theory-based approach to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. This study outlines an approach that may also be applicable to the development of interventions for other ethnic minority populations in diverse settings. There is now a need to build on this formative work and undertake a feasibility trial of a text message–based diet and physical activity intervention to prevent T2DM for women of Pakistani origin living in Scotland.
- Open Access
The epidemiological characteristics of stroke phenotypes defined with ICD-10 and free-text: a cohort study linked to electronic health records
Background Coded healthcare data may not capture all stroke cases and has limited accuracy for stroke subtypes. We sought to determine the incremental value of adding natural language processing (NLP) of free-text radiology reports to international classification of disease (ICD-10) codes to phenotype stroke, and stroke subtypes, in routinely collected healthcare datasets. Methods We linked participants in a community-based prospective cohort study, Generation Scotland, to clinical brain imaging reports (2008-2020) from five Scottish health boards. We used five combinations of NLP outputs and ICD-10 codes to define stroke phenotypes. With these phenotype models we measured the: stroke incidence standardised to a European Standardised Population; adjusted hazard ratio (aHR) of baseline hypertension for later stroke; and proportion of participants allocated stroke subtypes. Results Of 19,026 participants, over a mean follow-up of 10.2 years, 1938 had 3493 brain scans. Any stroke was identified in 534 participants: 319 with NLP alone, 59 with ICD-10 codes alone and 156 with both ICD-10 codes and an NLP report consistent with stroke. The stroke aHR for baseline hypertension was 1.47 (95%CI: 1.12-1.92) for NLP-defined stroke only; 1.57 (95%CI: 1.18-2.10) for ICD-10 defined stroke only; and 1.81 (95%CI: 1.20-2.72) for cases with ICD 10 stroke codes and NLP stroke phenotypes. The age-standardised incidence of stroke for these phenotype models was 1.35, 1.34, and 0.65 per 1000 person years, respectively. The proportion of strokes not subtyped was 26% (57/215) using only ICD-10, 9% (42/467) using only NLP, and 12% (65/534) using both NLP and ICD-10. Conclusions Addition of NLP derived phenotypes to ICD-10 stroke codes identified approximately 2.5 times more stroke cases and greatly increased the proportion with subtyping. The phenotype model using ICD 10 stroke codes and NLP stroke phenotypes had the strongest association with baseline hypertension. This information is relevant to large cohort studies and clinical trials that use routine electronic health records for outcome ascertainment.
Open AccessThe epidemiological characteristics of stroke phenotypes defined with ICD-10 and free-text: a cohort study linked to electronic health records
Background Coded healthcare data may not capture all stroke cases and has limited accuracy for stroke subtypes. We sought to determine the incremental value of adding natural language processing (NLP) of free-text radiology reports to international classification of disease (ICD-10) codes to phenotype stroke, and stroke subtypes, in routinely collected healthcare datasets. Methods We linked participants in a community-based prospective cohort study, Generation Scotland, to clinical brain imaging reports (2008-2020) from five Scottish health boards. We used five combinations of NLP outputs and ICD-10 codes to define stroke phenotypes. With these phenotype models we measured the: stroke incidence standardised to a European Standardised Population; adjusted hazard ratio (aHR) of baseline hypertension for later stroke; and proportion of participants allocated stroke subtypes. Results Of 19,026 participants, over a mean follow-up of 10.2 years, 1938 had 3493 brain scans. Any stroke was identified in 534 participants: 319 with NLP alone, 59 with ICD-10 codes alone and 156 with both ICD-10 codes and an NLP report consistent with stroke. The stroke aHR for baseline hypertension was 1.47 (95%CI: 1.12-1.92) for NLP-defined stroke only; 1.57 (95%CI: 1.18-2.10) for ICD-10 defined stroke only; and 1.81 (95%CI: 1.20-2.72) for cases with ICD 10 stroke codes and NLP stroke phenotypes. The age-standardised incidence of stroke for these phenotype models was 1.35, 1.34, and 0.65 per 1000 person years, respectively. The proportion of strokes not subtyped was 26% (57/215) using only ICD-10, 9% (42/467) using only NLP, and 12% (65/534) using both NLP and ICD-10. Conclusions Addition of NLP derived phenotypes to ICD-10 stroke codes identified approximately 2.5 times more stroke cases and greatly increased the proportion with subtyping. The phenotype model using ICD 10 stroke codes and NLP stroke phenotypes had the strongest association with baseline hypertension. This information is relevant to large cohort studies and clinical trials that use routine electronic health records for outcome ascertainment.
- Open Access
- 2 October 2021
- Published by Springer Nature in BMC Medical Imaging
- Vol. 21 (1) , 1-13
The reporting quality of natural language processing studies: systematic review of studies of radiology reports
Background: Automated language analysis of radiology reports using natural language processing (NLP) can provide valuable information on patients’ health and disease. With its rapid development, NLP studies should have transparent methodology to allow comparison of approaches and reproducibility. This systematic review aims to summarise the characteristics and reporting quality of studies applying NLP to radiology reports. Methods: We searched Google Scholar for studies published in English that applied NLP to radiology reports of any imaging modality between January 2015 and October 2019. At least two reviewers independently performed screening and completed data extraction. We specified 15 criteria relating to data source, datasets, ground truth, outcomes, and reproducibility for quality assessment. The primary NLP performance measures were precision, recall and F1 score. Results: Of the 4,836 records retrieved, we included 164 studies that used NLP on radiology reports. The commonest clinical applications of NLP were disease information or classification (28%) and diagnostic surveillance (27.4%). Most studies used English radiology reports (86%). Reports from mixed imaging modalities were used in 28% of the studies. Oncology (24%) was the most frequent disease area. Most studies had dataset size > 200 (85.4%) but the proportion of studies that described their annotated, training, validation, and test set were 67.1%, 63.4%, 45.7%, and 67.7% respectively. About half of the studies reported precision (48.8%) and recall (53.7%). Few studies reported external validation performed (10.8%), data availability (8.5%) and code availability (9.1%). There was no pattern of performance associated with the overall reporting quality. Conclusions: There is a range of potential clinical applications for NLP of radiology reports in health services and research. However, we found suboptimal reporting quality that precludes comparison, reproducibility, and replication. Our results support the need for development of reporting standards specific to clinical NLP studies.
- 2 October 2021
- Published by Springer Nature in BMC Medical Imaging
- Vol. 21 (1) , 1-13
Open AccessThe reporting quality of natural language processing studies: systematic review of studies of radiology reports
Background: Automated language analysis of radiology reports using natural language processing (NLP) can provide valuable information on patients’ health and disease. With its rapid development, NLP studies should have transparent methodology to allow comparison of approaches and reproducibility. This systematic review aims to summarise the characteristics and reporting quality of studies applying NLP to radiology reports. Methods: We searched Google Scholar for studies published in English that applied NLP to radiology reports of any imaging modality between January 2015 and October 2019. At least two reviewers independently performed screening and completed data extraction. We specified 15 criteria relating to data source, datasets, ground truth, outcomes, and reproducibility for quality assessment. The primary NLP performance measures were precision, recall and F1 score. Results: Of the 4,836 records retrieved, we included 164 studies that used NLP on radiology reports. The commonest clinical applications of NLP were disease information or classification (28%) and diagnostic surveillance (27.4%). Most studies used English radiology reports (86%). Reports from mixed imaging modalities were used in 28% of the studies. Oncology (24%) was the most frequent disease area. Most studies had dataset size > 200 (85.4%) but the proportion of studies that described their annotated, training, validation, and test set were 67.1%, 63.4%, 45.7%, and 67.7% respectively. About half of the studies reported precision (48.8%) and recall (53.7%). Few studies reported external validation performed (10.8%), data availability (8.5%) and code availability (9.1%). There was no pattern of performance associated with the overall reporting quality. Conclusions: There is a range of potential clinical applications for NLP of radiology reports in health services and research. However, we found suboptimal reporting quality that precludes comparison, reproducibility, and replication. Our results support the need for development of reporting standards specific to clinical NLP studies.
- Open Access
- 24 September 2021
- Published by JMIR Publications Inc. in JMIR Preprints
A Culturally Adapted Diet and Physical Activity Text Message Intervention to Prevent Type 2 Diabetes Mellitus for Women of Pakistani Origin Living in Scotland: Formative Study (Preprint)
BACKGROUND Individuals of South Asian origin are at an increased risk of developing type 2 diabetes mellitus (T2DM) compared with other ethnic minority groups. Therefore, there is a need to develop interventions to address, and reduce, this heightened risk. OBJECTIVE We undertook formative work to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. METHODS We used a stepwise approach that was informed by the Six Steps in Quality Intervention Development framework, which consisted of gathering evidence through literature review and focus groups (step 1), developing a program theory for the intervention (step 2), and finally developing the content of the text messages and an accompanying delivery plan (step 3). RESULTS In step 1, we reviewed 12 articles and identified 3 key themes describing factors impacting on diet and physical activity in the context of T2DM prevention: knowledge on ways to prevent T2DM through diet and physical activity; cultural, social, and gender norms; and perceived level of control and sense of inevitability over developing T2DM. The key themes that emerged from the 3 focus groups with a total of 25 women were the need for interventions to provide “friendly encouragement,” “companionship,” and a “focus on the individual” and also for the text messages to “set achievable goals” and include “information on cooking healthy meals.” We combined the findings of the focus groups and literature review to create 13 guiding principles for culturally adapting the text messages. In step 2, we developed a program theory, which specified the main determinants of change that our text messages should aim to enhance: knowledge and skills, sense of control, goal setting and planning behavior, peer support, and norms and beliefs guiding behavior. In step 3, we used both the intervention program theory and guiding principles to develop a set of 73 text messages aimed at supporting a healthy diet and 65 text messages supporting increasing physical activity. CONCLUSIONS We present a theory-based approach to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. This study outlines an approach that may also be applicable to the development of interventions for other ethnic minority populations in diverse settings. There is now a need to build on this formative work and undertake a feasibility trial of a text message–based diet and physical activity intervention to prevent T2DM for women of Pakistani origin living in Scotland.
- 24 September 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessA Culturally Adapted Diet and Physical Activity Text Message Intervention to Prevent Type 2 Diabetes Mellitus for Women of Pakistani Origin Living in Scotland: Formative Study (Preprint)
BACKGROUND Individuals of South Asian origin are at an increased risk of developing type 2 diabetes mellitus (T2DM) compared with other ethnic minority groups. Therefore, there is a need to develop interventions to address, and reduce, this heightened risk. OBJECTIVE We undertook formative work to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. METHODS We used a stepwise approach that was informed by the Six Steps in Quality Intervention Development framework, which consisted of gathering evidence through literature review and focus groups (step 1), developing a program theory for the intervention (step 2), and finally developing the content of the text messages and an accompanying delivery plan (step 3). RESULTS In step 1, we reviewed 12 articles and identified 3 key themes describing factors impacting on diet and physical activity in the context of T2DM prevention: knowledge on ways to prevent T2DM through diet and physical activity; cultural, social, and gender norms; and perceived level of control and sense of inevitability over developing T2DM. The key themes that emerged from the 3 focus groups with a total of 25 women were the need for interventions to provide “friendly encouragement,” “companionship,” and a “focus on the individual” and also for the text messages to “set achievable goals” and include “information on cooking healthy meals.” We combined the findings of the focus groups and literature review to create 13 guiding principles for culturally adapting the text messages. In step 2, we developed a program theory, which specified the main determinants of change that our text messages should aim to enhance: knowledge and skills, sense of control, goal setting and planning behavior, peer support, and norms and beliefs guiding behavior. In step 3, we used both the intervention program theory and guiding principles to develop a set of 73 text messages aimed at supporting a healthy diet and 65 text messages supporting increasing physical activity. CONCLUSIONS We present a theory-based approach to develop a culturally adapted diet and physical activity text message intervention to prevent T2DM for women of Pakistani origin living in Scotland. This study outlines an approach that may also be applicable to the development of interventions for other ethnic minority populations in diverse settings. There is now a need to build on this formative work and undertake a feasibility trial of a text message–based diet and physical activity intervention to prevent T2DM for women of Pakistani origin living in Scotland.
- Top Cited PapersOpen Access
- 3 June 2021
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 21 (1) , 1-18
A systematic review of natural language processing applied to radiology reports
Background: Natural language processing (NLP) has a significant role in advancing healthcare and has been found to be key in extracting structured information from radiology reports. Understanding recent developments in NLP application to radiology is of significance but recent reviews on this are limited. This study systematically assesses and quantifies recent literature in NLP applied to radiology reports.Methods: We conduct an automated literature search yielding 4836 results using automated filtering, metadata enriching steps and citation search combined with manual review. Our analysis is based on 21 variables including radiology characteristics, NLP methodology, performance, study, and clinical application characteristics.Results: We present a comprehensive analysis of the 164 publications retrieved with publications in 2019 almost triple those in 2015. Each publication is categorised into one of 6 clinical application categories. Deep learning use increases in the period but conventional machine learning approaches are still prevalent. Deep learning remains challenged when data is scarce and there is little evidence of adoption into clinical practice. Despite 17% of studies reporting greater than 0.85 F1 scores, it is hard to comparatively evaluate these approaches given that most of them use different datasets. Only 14 studies made their data and 15 their code available with 10 externally validating results.Conclusions: Automated understanding of clinical narratives of the radiology reports has the potential to enhance the healthcare process and we show that research in this field continues to grow. Reproducibility and explainability of models are important if the domain is to move applications into clinical use. More could be done to share code enabling validation of methods on different institutional data and to reduce heterogeneity in reporting of study properties allowing inter-study comparisons. Our results have significance for researchers in the field providing a systematic synthesis of existing work to build on, identify gaps, opportunities for collaboration and avoid duplication.
- 3 June 2021
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 21 (1) , 1-18
Top Cited PapersOpen AccessA systematic review of natural language processing applied to radiology reports
Background: Natural language processing (NLP) has a significant role in advancing healthcare and has been found to be key in extracting structured information from radiology reports. Understanding recent developments in NLP application to radiology is of significance but recent reviews on this are limited. This study systematically assesses and quantifies recent literature in NLP applied to radiology reports.Methods: We conduct an automated literature search yielding 4836 results using automated filtering, metadata enriching steps and citation search combined with manual review. Our analysis is based on 21 variables including radiology characteristics, NLP methodology, performance, study, and clinical application characteristics.Results: We present a comprehensive analysis of the 164 publications retrieved with publications in 2019 almost triple those in 2015. Each publication is categorised into one of 6 clinical application categories. Deep learning use increases in the period but conventional machine learning approaches are still prevalent. Deep learning remains challenged when data is scarce and there is little evidence of adoption into clinical practice. Despite 17% of studies reporting greater than 0.85 F1 scores, it is hard to comparatively evaluate these approaches given that most of them use different datasets. Only 14 studies made their data and 15 their code available with 10 externally validating results.Conclusions: Automated understanding of clinical narratives of the radiology reports has the potential to enhance the healthcare process and we show that research in this field continues to grow. Reproducibility and explainability of models are important if the domain is to move applications into clinical use. More could be done to share code enabling validation of methods on different institutional data and to reduce heterogeneity in reporting of study properties allowing inter-study comparisons. Our results have significance for researchers in the field providing a systematic synthesis of existing work to build on, identify gaps, opportunities for collaboration and avoid duplication.
- Open Access
- 10 April 2021
- Published by Wiley in Diabetic Medicine
- Vol. 38 (11) , e14584
Developing a realist informed framework for cultural adaptation of lifestyle interventions for the prevention of type 2 diabetes in South Asian populations in Europe
Aims: Selected lifestyle interventions proven effective for White‐European populations have been culturally adapted for South Asian populations living in Europe, who are at higher risk of type 2 diabetes. However, a limited theoretical basis underpins how cultural adaptations are believed to augment intervention effectiveness. We undertook a realist review to synthesise existing literature on culturally adapted type 2 diabetes prevention interventions, to develop a framework that shows ‘how’ cultural adaptation works, for ‘whom’ and in ‘what contexts’.Methods: We followed the stepped methodological approach of realist review. Our work concluded a European‐wide project (EuroDHYAN), and core studies were identified from the preceding EuroDHYAN reviews. Data were extracted, coded into themes and synthesised to create ‘Context–Mechanism–Outcome’ configurations and to generate a refined explanatory framework.Results: We identified eight core intervention papers. From this evidence, and supporting literature, we examined the ‘Team’ domain of cultural adaptation and identified a mechanism of shared cultural identity which we theorised as contributing to strong team‐participant relationships. We also identified four key contexts which influenced intervention outcomes: ‘research setting’ and ‘heterogeneous populations’ (intrinsic to the intervention) and ‘broader environment’ and ‘socio‐cultural stress’ (extrinsic barriers).Conclusions: This work instigates research into the mechanisms of cultural adaptation which, if pursued, will allow a more nuanced understanding of how to apply adaptations, and for whom. In practice we recommend greater consideration of heterogeneous and intersecting population characteristics; how intervention design can safeguard sustainability; and how the four key contexts identified influence how, and whether, these interventions work.
- 10 April 2021
- Published by Wiley in Diabetic Medicine
- Vol. 38 (11) , e14584
Open AccessDeveloping a realist informed framework for cultural adaptation of lifestyle interventions for the prevention of type 2 diabetes in South Asian populations in Europe
Aims: Selected lifestyle interventions proven effective for White‐European populations have been culturally adapted for South Asian populations living in Europe, who are at higher risk of type 2 diabetes. However, a limited theoretical basis underpins how cultural adaptations are believed to augment intervention effectiveness. We undertook a realist review to synthesise existing literature on culturally adapted type 2 diabetes prevention interventions, to develop a framework that shows ‘how’ cultural adaptation works, for ‘whom’ and in ‘what contexts’.Methods: We followed the stepped methodological approach of realist review. Our work concluded a European‐wide project (EuroDHYAN), and core studies were identified from the preceding EuroDHYAN reviews. Data were extracted, coded into themes and synthesised to create ‘Context–Mechanism–Outcome’ configurations and to generate a refined explanatory framework.Results: We identified eight core intervention papers. From this evidence, and supporting literature, we examined the ‘Team’ domain of cultural adaptation and identified a mechanism of shared cultural identity which we theorised as contributing to strong team‐participant relationships. We also identified four key contexts which influenced intervention outcomes: ‘research setting’ and ‘heterogeneous populations’ (intrinsic to the intervention) and ‘broader environment’ and ‘socio‐cultural stress’ (extrinsic barriers).Conclusions: This work instigates research into the mechanisms of cultural adaptation which, if pursued, will allow a more nuanced understanding of how to apply adaptations, and for whom. In practice we recommend greater consideration of heterogeneous and intersecting population characteristics; how intervention design can safeguard sustainability; and how the four key contexts identified influence how, and whether, these interventions work.
- Open Access
- 10 December 2020
- Published by Oxford University Press (OUP) in Journal of Public Health
- Vol. 43 (4) , e728-e738
Raising ethnicity recording in NHS Lothian from 3% to 90% in 3 years: processes and analysis of data from Accidents and Emergencies
Background: The disproportionate burden of COVID-19 on ethnic minority populations has recently highlighted the necessity of maintaining accessible, routinely collected, ethnicity data within healthcare services. Despite 25 years of supportive legislation and policy in the UK, ethnicity data recording remains inconsistent, which has hindered needs assessment, evaluation and decision-making. We describe efforts to improve the completeness, quality and usage of ethnicity data within our regional health board, NHS Lothian. Methods: The Ethnicity Coding Task Force was established with the aim of increasing ethnicity recording within NHS Lothian secondary care services from 3 to 90% over 3 years. We subsequently analysed these data specifically focusing on Accident and Emergency (A&E) use by ethnic group. Results: We achieved 91%, 85% and 93% completeness of recording across inpatients, outpatients and A&E, respectively. Analysis of A&E data found a mixed pattern of attendance amongst ethnic minority populations and did not support the commonly perceived relationship between lower GP registration and higher A&E use within this population. Conclusions: We identified a successful approach to increase ethnicity recording within a regional health board, which could potentially be useful in other settings, and demonstrated the utility of these data in informing assessment of healthcare delivery and future planning.
- 10 December 2020
- Published by Oxford University Press (OUP) in Journal of Public Health
- Vol. 43 (4) , e728-e738
Open AccessRaising ethnicity recording in NHS Lothian from 3% to 90% in 3 years: processes and analysis of data from Accidents and Emergencies
Background: The disproportionate burden of COVID-19 on ethnic minority populations has recently highlighted the necessity of maintaining accessible, routinely collected, ethnicity data within healthcare services. Despite 25 years of supportive legislation and policy in the UK, ethnicity data recording remains inconsistent, which has hindered needs assessment, evaluation and decision-making. We describe efforts to improve the completeness, quality and usage of ethnicity data within our regional health board, NHS Lothian. Methods: The Ethnicity Coding Task Force was established with the aim of increasing ethnicity recording within NHS Lothian secondary care services from 3 to 90% over 3 years. We subsequently analysed these data specifically focusing on Accident and Emergency (A&E) use by ethnic group. Results: We achieved 91%, 85% and 93% completeness of recording across inpatients, outpatients and A&E, respectively. Analysis of A&E data found a mixed pattern of attendance amongst ethnic minority populations and did not support the commonly perceived relationship between lower GP registration and higher A&E use within this population. Conclusions: We identified a successful approach to increase ethnicity recording within a regional health board, which could potentially be useful in other settings, and demonstrated the utility of these data in informing assessment of healthcare delivery and future planning.
- 22 November 2020
- Published by Wiley in Australian Occupational Therapy Journal
- Vol. 68 (2) , 144-155
The Power of Playgroups: Key components of supported and therapeutic playgroups from the perspective of parents
Introduction Playgroups are community‐based programs for children and families aiming to improve child outcomes, enhance family and community networks and increase parenting capacity. Despite the prevalence of playgroups in Australian communities there is a lack of research clearly articulating the key components of playgroups, specifically from the perspective of parents attending these groups. This study aimed to identify the key components of supported and therapeutic playgroups impacting on perceived effectiveness from the perspective of parents with a child with a developmental delay and/or disability. Methods This study explored the experiences of 23 parents attending supported or therapeutic playgroups using a qualitative interpretive phenomenological approach. Data were collected through three focus groups and seven individual interviews and analysed using Colaizzi's (1978) qualitative method of data analysis. Results Findings indicated playgroup components that most strongly impacted on perceived effectiveness were feeling accepted; providing opportunities for child development, socialisation and enjoyment; and enhancing parental knowledge and skills. Findings reinforced the importance of family centred practice and facilitating peer support for families of children with developmental delay and/or disability. Conclusion Supported and therapeutic playgroups emerged as a valuable model for parents of children with developmental delays and/or disabilities but require an interplay of specific facilitator, parent and child characteristics to be effective. This study contributes to the understanding of key components of successful supported and therapeutic playgroup models, highlighting the importance of engaging consumers in developing evidence‐based meaningful interventions for children with developmental delay and/or disabilities and their families.
- 22 November 2020
- Published by Wiley in Australian Occupational Therapy Journal
- Vol. 68 (2) , 144-155
The Power of Playgroups: Key components of supported and therapeutic playgroups from the perspective of parents
Introduction Playgroups are community‐based programs for children and families aiming to improve child outcomes, enhance family and community networks and increase parenting capacity. Despite the prevalence of playgroups in Australian communities there is a lack of research clearly articulating the key components of playgroups, specifically from the perspective of parents attending these groups. This study aimed to identify the key components of supported and therapeutic playgroups impacting on perceived effectiveness from the perspective of parents with a child with a developmental delay and/or disability. Methods This study explored the experiences of 23 parents attending supported or therapeutic playgroups using a qualitative interpretive phenomenological approach. Data were collected through three focus groups and seven individual interviews and analysed using Colaizzi's (1978) qualitative method of data analysis. Results Findings indicated playgroup components that most strongly impacted on perceived effectiveness were feeling accepted; providing opportunities for child development, socialisation and enjoyment; and enhancing parental knowledge and skills. Findings reinforced the importance of family centred practice and facilitating peer support for families of children with developmental delay and/or disability. Conclusion Supported and therapeutic playgroups emerged as a valuable model for parents of children with developmental delays and/or disabilities but require an interplay of specific facilitator, parent and child characteristics to be effective. This study contributes to the understanding of key components of successful supported and therapeutic playgroup models, highlighting the importance of engaging consumers in developing evidence‐based meaningful interventions for children with developmental delay and/or disabilities and their families.
- Open Access
- 1 August 2018
- Published by Elsevier in Public Health
- Vol. 161, 50-58
Developing culturally adapted lifestyle interventions for South Asian migrant populations: a qualitative study of the key success factors and main challenges
- 1 August 2018
- Published by Elsevier in Public Health
- Vol. 161, 50-58
Open AccessDeveloping culturally adapted lifestyle interventions for South Asian migrant populations: a qualitative study of the key success factors and main challenges