• Digital Humanities and Cultural Heritage
• Open Access close
• Publications close
• Research software close
• Other research products close
• Wellcome Trust close
• The Alan Turing Institute close
• Digital Humanities and Cultural Her... close

Background: With the improvements to text mining technology and the availability of large unstructured Electronic Healthcare Records (EHR) datasets, it is now possible to extract structured information from raw text contained within EHR at reasonably high accuracy. We describe a text mining system for classifying radiologists’ reports of CT and MRI brain scans, assigning labels indicating occurrence and type of stroke, as well as other observations. Our system, the Edinburgh Information Extraction for Radiology reports (EdIE-R) system, which we describe here, was developed and tested on a collection of radiology reports. Methods: The work reported in this paper is based on 1,168 radiology reports from the Edinburgh Stroke Study (ESS) [1], a hospital-based register of stroke and transient ischaemic attack patients. We created manually annotations for this data in parallel with developing the rule-based EdIE-R system to identify phenotype information related to stroke in radiology reports. This process was iterative and domain expert feedback was considered at each iteration to adapt and tune the EdIE-R text mining system which identifies entities, negation and relations between entities in each report and determines report-level labels. Results:The inter-annotator agreement (IAA) for all types of annotations is high at 96.96 for entities, 96.46 for negation, 95.84 for relations and 94.02 for labels. The equivalent system scores on the blind test set are equally high at 95.49 for entities, 94.41 for negation, 98.27 for relations and 96.39 for labels for the first annotator and 96.86, 96.01, 96.53 and 92.61, respectively for the second annotator. Conclusion: Automated reading of such EHR data at such high levels of accuracies opens up avenues for population health monitoring and audit, and can provide a resource for epidemiological studies. We are in the process of validating EdIE-R in separate larger cohorts in NHS England and Scotland. The manually annotated ESS corpus will be available for research purposes on application.

Abstract Background Manual coding of phenotypes in brain radiology reports is time consuming. We developed a natural language processing (NLP) algorithm to enable automatic identification of brain imaging in radiology reports performed in routine clinical practice in the UK National Health Service (NHS). Methods We used anonymized text brain imaging reports from a cohort study of stroke/TIA patients and from a regional hospital to develop and test an NLP algorithm. Two experts marked up text in 1692 reports for 24 cerebrovascular and other neurological phenotypes. We developed and tested a rule-based NLP algorithm first within the cohort study, and further evaluated it in the reports from the regional hospital. Results The agreement between expert readers was excellent (Cohen’s κ =0.93) in both datasets. In the final test dataset (n = 700) in unseen regional hospital reports, the algorithm had very good performance for a report of any ischaemic stroke [sensitivity 89% (95% CI:81–94); positive predictive value (PPV) 85% (76–90); specificity 100% (95% CI:0.99–1.00)]; any haemorrhagic stroke [sensitivity 96% (95% CI: 80–99), PPV 72% (95% CI:55–84); specificity 100% (95% CI:0.99–1.00)]; brain tumours [sensitivity 96% (CI:87–99); PPV 84% (73–91); specificity: 100% (95% CI:0.99–1.00)] and cerebral small vessel disease and cerebral atrophy (sensitivity, PPV and specificity all > 97%). We obtained few reports of subarachnoid haemorrhage, microbleeds or subdural haematomas. In 110,695 reports from NHS Tayside, atrophy (n = 28,757, 26%), small vessel disease (15,015, 14%) and old, deep ischaemic strokes (10,636, 10%) were the commonest findings. Conclusions An NLP algorithm can be developed in UK NHS radiology records to allow identification of cohorts of patients with important brain imaging phenotypes at a scale that would otherwise not be possible.

Abstract Deep convolutional neural networks (DCNNs) rival humans in object recognition. The layers (or levels of representation) in DCNNs have been successfully aligned with processing stages along the ventral stream for visual processing. Here, we propose a model of concept learning that uses visual representations from these networks to build memory representations of novel categories, which may rely on the medial temporal lobe (MTL) and medial prefrontal cortex (mPFC). Our approach opens up two possibilities: a) formal investigations can involve photographic stimuli as opposed to stimuli handcrafted and coded by the experimenter; b) model comparison can determine which level of representation within a DCNN a learner is using during categorization decisions. Pursuing the latter point, DCNNs suggest that the shape bias in children relies on representations at more advanced network layers whereas a learner that relied on lower network layers would display a color bias. These results confirm the role of natural statistics in the shape bias (i.e., shape is predictive of category membership) while highlighting that the type of statistics matter, i.e., those from lower or higher levels of representation. We use the same approach to provide evidence that pigeons performing seemingly sophisticated categorization of complex imagery may in fact be relying on representations that are very low-level (i.e., retinotopic). Although complex features, such as shape, relatively predominate at more advanced network layers, even simple features, such as spatial frequency and orientation, are better represented at the more advanced layers, contrary to a standard hierarchical view. Author Note This work was supported by NIH (Grant 1P01HD080679), and a Wellcome Trust Investigator Award (Grant WT106931MA) to BCL, as well as The Alan Turing Institute under the EPSRC grant EP/N510129/1. Some of this work was originally reported at the 39th Annual Meeting of the Cognitive Science Society in 2017. Correspondences regarding this work can be sent to o.guest@ucl.ac.uk. The authors declare that they have no competing interests. The authors would like to thank the members of the Love lab at UCL for their useful input when preparing the manuscript. The code used to run these experiments is available here: https://osf.io/jxavn/.

This manuscript was written as a result of a workshop that was held at the Institute of Psychiatry, Psychology and Neuroscience, King’s College London, financially supported by the European Science Foundation (ESF) Research Networking Programme Evaluating Information Access Systems: http://elias-network.eu/. SV is supported by the Swedish Research Council (2015-00359) and the Marie Skłodowska Curie Actions, Cofund, Project INCA 600398. EB-G is partially supported by grants from Instituto de Salud Carlos III (ISCIII PI13/02200; PI16/01852), Delegación del Gobierno para el Plan Nacional de Drogas (20151073); American Foundation for Suicide Prevention (AFSP) (LSRG- 1-005-16). NW is supported by the UCLH NIHR Biomedical Research Centre. DN is supported by The Alan Turing Institute under the EPSRC grant EP/N510129/1, with an Alan Turing Institute Fellowship (TU/A/000006). RP has received support from a Medical Research Council (MRC) Health Data Research UK Fellowship (MR/S003118/1) and a Starter Grant for Clinical Lecturers (SGL015/1020) supported by the Academy of Medical Sciences, The Wellcome Trust, MRC, British Heart Foundation, Arthritis Research UK, the Royal College of Physicians and Diabetes UK. DL is supported by the UK Medical Research Council under grant MR/N028244/2 and the King’s Centre for Military Health Research. JD is supported by a Medical Research Council (MRC) Clinical Research Training Fellowship (MR/L017105/1). RD is funded by a Clinician Scientist Fellowship (research project e-HOST-IT) from the Health Foundation in partnership with the Academy of Medical Sciences Risk assessment of suicidal behavior is a time-consuming but notoriously inaccurate activity formental health services globally. In the last 50 years a large number of tools have been designed for suicide risk assessment, and tested in a wide variety of populations, but studies show that these tools suffer fromlow positive predictive values.More recently, advances in research fields such as machine learning and natural language processing applied on large datasets have shown promising results for health care, and may enable an important shift in advancing precision medicine. In this conceptual review, we discuss established risk assessment tools and examples of novel data-driven approaches that have been used for identification of suicidal behavior and risk. We provide a perspective on the strengths and weaknesses of these applications to mental health-related data, and suggest research directions to enable improvement in clinical practice