Analysis of Stroke Detection during the COVID-19 Pandemic Using Natural Language Processing of Radiology Reports

BACKGROUND AND PURPOSE:The coronavirus disease 2019 (COVID-19) pandemic has led to decreases in neuroimaging volume. Our aim was to quantify the change in acute or subacute ischemic strokes detected on CT or MR imaging during the pandemic using natural language processing of radiology reports.MATERIALS AND METHODS:We retrospectively analyzed 32,555 radiology reports from brain CTs and MRIs from a comprehensive stroke center, performed from March 1 to April 30 each year from 2017 to 2020, involving 20,414 unique patients. To detect acute or subacute ischemic stroke in free-text reports, we trained a random forest natural language processing classifier using 1987 randomly sampled radiology reports with manual annotation. Natural language processing classifier generalizability was evaluated using 1974 imaging reports from an external dataset.RESULTS:The natural language processing classifier achieved a 5-fold cross-validation classification accuracy of 0.97 and an F1 score of 0.74, with a slight underestimation (−5%) of actual numbers of acute or subacute ischemic strokes in cross-validation. Importantly, cross-validation performance stratified by year was similar. Applying the classifier to the complete study cohort, we found an estimated 24% decrease in patients with acute or subacute ischemic strokes reported on CT or MR imaging from March to April 2020 compared with the average from those months in 2017–2019. Among patients with stroke-related order indications, the estimated proportion who underwent neuroimaging with acute or subacute ischemic stroke detection significantly increased from 16% during 2017–2019 to 21% in 2020 (P = .01). The natural language processing classifier performed worse on external data.CONCLUSIONS:Acute or subacute ischemic stroke cases detected by neuroimaging decreased during the COVID-19 pandemic, though a higher proportion of studies ordered for stroke were positive for acute or subacute ischemic strokes. Natural language processing approaches can help automatically track acute or subacute ischemic stroke numbers for epidemiologic studies, though local classifier training is important due to radiologist reporting style differences.

There is much concern regarding the impact of the coronavirus disease 2019 (COVID-19) pandemic on the quality of stroke care, including issues with hospital capacity, clinical resource re-allocation, and the safety of patients and clinicians.^1,2 Previous reports have shown that there have been substantial decreases in stroke neuroimaging volume during the pandemic.^3,4 In addition, acute ischemic infarcts have been found on neuroimaging studies in many hospitalized patients with COVID-19, though the causal relationship is unclear.^5,6 Studies like these and other epidemiologic analyses usually rely on the creation of manually curated databases, in which identification of cases can be time-consuming and difficult to update in real-time. One way to facilitate such research is to use natural language processing (NLP), which has shown utility for automated analysis of radiology report data.⁷ NLP algorithms have been developed previously for the classification of neuroradiology reports for the presence of ischemic stroke findings and acute ischemic stroke subtypes.^8,9 Thus, NLP has the potential to facilitate COVID-19 research.In this study, we developed an NLP machine learning model that classifies radiology reports for the presence or absence of acute or subacute ischemic stroke (ASIS), as opposed to chronic stroke. We used this model to quantify the change in ASIS detected on all CT or MR imaging studies performed at a large comprehensive stroke center during the COVID-19 pandemic in the United States. We also evaluated NLP model generalizability and different training strategies using a sample of radiology reports from a second stroke center.     

Barriers and facilitators for cascade testing in genetic conditions: a systematic review

Cascade testing is the process of offering genetic counseling and testing to at-risk relatives of an individual who has been diagnosed with a genetic condition. It is critical for increasing the identification rates of individuals with these conditions and the uptake of appropriate preventive health services. The process of cascade testing is highly varied in clinical practice, and a comprehensive understanding of factors that hinder or enhance its implementation is necessary to improve this process. We conducted a systematic review to identify barriers and facilitators for cascade testing and searched PubMed, CINAHL via EBSCO, Web of Science, EMBASE, and the Cochrane Library for articles published from the databases’ inception to November 2018. Thirty articles met inclusion criteria. Barriers and facilitators identified from these studies at the individual-level were organized into the following categories: (1) demographics, (2) knowledge, (3) attitudes, beliefs, and emotional responses of the individual, and (4) perceptions of relatives, relatives’ responses, and attitudes toward relatives. At the interpersonal-level, barriers and facilitators were categorized as (1) family communication-, support- and dynamics-, and (2) provider-factors. Finally, barriers at the environmental-level relating to accessibility of genetic services were also identified. Our findings suggest that several individual, interpersonal and environmental factors may play a role in cascade testing. Future studies to further investigate these barriers and facilitators are needed to inform future interventions for improving the implementation of cascade testing for genetic conditions in clinical practice.Subject terms: Preventive medicine, Genetic services     

Shared decision-making for the management of small renal masses: Development and acceptability testing of a novel patient decision aid

IntroductionShared decision-making incorporates patients’ values and preferences to achieve high-quality decisions. The objective of this study was to develop an acceptable patient decision aid to facilitate shared decision-making for the management of small renal masses (SRMs).MethodsThe International Patient Decision Aids Standards were used to guide an evidence-based development process. Management options included active surveillance, thermal ablation, partial nephrectomy, and radical nephrectomy. A literature review was performed to provide incidence rates for outcomes of each option. Once a prototype was complete, alpha-testing was performed using a 10-question survey to assess acceptability with patients, patient advocates, urologists, and methodological experts. The primary outcome was acceptability of the decision aid.ResultsA novel patient decision aid was created to facilitate shared decision-making for the management of SRMs. Acceptability testing was performed with 20 patients, 10 urologists, two patient advocates, and one methodological expert. Responders indicated the decision aid was appropriate in length (82%, 27/33), well-balanced (82%, 27/33), and had language that was easy to follow (94%, 31/33). All patient responders felt the decision aid would have been helpful during their consultation and would recommend the decision aid for future patients (100%, 20/20). Most urologists reported they intend to use the decision aid (90%, 9/10).ConclusionsA novel patient decision aid was created to facilitate shared decision-making for management of SRMs. This clinical tool was acceptable with patients, patient advocates, and urologists and is freely available at: https://decisionaid.ohri.ca/decaids.html.     

An uncertainty principle for neural coding: Conjugate representations of position and velocity are mapped onto firing rates and co‐firing rates of neural spike trains

The hippocampal system contains neural populations that encode an animal's position and velocity as it navigates through space. Here, we show that such populations can embed two codes within their spike trains: a firing rate code ( R ) conveyed by within‐cell spike intervals, and a co‐firing rate code () conveyed by between‐cell spike intervals. These two codes behave as conjugates of one another, obeying an analog of the uncertainty principle from physics: information conveyed in R comes at the expense of information in , and vice versa. An exception to this trade‐off occurs when spike trains encode a pair of conjugate variables, such as position and velocity, which do not compete for capacity across R and . To illustrate this, we describe two biologically inspired methods for decoding R and , referred to as sigma and sigma‐chi decoding, respectively. Simulations of head direction and grid cells show that if firing rates are tuned for position (but not velocity), then position is recovered by sigma decoding, whereas velocity is recovered by sigma‐chi decoding. Conversely, simulations of oscillatory interference among theta‐modulated “speed cells” show that if co‐firing rates are tuned for position (but not velocity), then position is recovered by sigma‐chi decoding, whereas velocity is recovered by sigma decoding. Between these two extremes, information about both variables can be distributed across both channels, and partially recovered by both decoders. These results suggest that populations with different spatial and temporal tuning properties—such as speed versus grid cells—might not encode different information, but rather, distribute similar information about position and velocity in different ways across R and . Such conjugate coding of position and velocity may influence how hippocampal populations are interconnected to form functional circuits, and how biological neurons integrate their inputs to decode information from firing rates and spike correlations.     

Managing invasive aspergillosis in haematological patients in the era of resistance polymerase chain reaction and increasing triazole resistance: A modelling study of different strategies

Objectives

Triazole resistance in Aspergillus spp. is emerging and complicates prophylaxis and treatment of invasive aspergillosis (IA) worldwide. New polymerase chain reaction (PCR) tests on broncho-alveolar lavage (BAL) fluid allow for detection of triazole resistance at a genetic level, which has opened up new possibilities for targeted therapy. In the absence of clinical trials, a modelling study delivers estimates of the added value of resistance detection with PCR, and which empiric therapy would be optimal when local resistance rates are known.

Positive association between cerebral grey matter metabolism and dopamine D2/D3 receptor availability in healthy and schizophrenia subjects: An 18F-fluorodeoxyglucose and 18F-fallypride positron emission tomography study

Abstract

Objectives: Overlapping decreases in extrastriatal dopamine D₂/D₃-receptor availability and glucose metabolism have been reported in subjects with schizophrenia. It remains unknown whether these findings are physiologically related or coincidental.

Methods: To ascertain this, we used two consecutive ¹⁸F-fluorodeoxyglucose and ¹⁸F-fallypride positron emission tomography scans in 19 healthy and 25 unmedicated schizophrenia subjects. Matrices of correlations between ¹⁸F-fluorodeoxyglucose uptake and ¹⁸F-fallypride binding in voxels at the same xyz location and AFNI-generated regions of interest were evaluated in both diagnostic groups.

Results: ¹⁸F-fluorodeoxyglucose uptake and ¹⁸F-fallypride binding potential were predominantly positively correlated across the striatal and extrastriatal grey matter in both healthy and schizophrenia subjects. In comparison to healthy subjects, significantly weaker correlations in subjects with schizophrenia were confirmed in the right cingulate gyrus and thalamus, including the mediodorsal, lateral dorsal, anterior, and midline nuclei. Schizophrenia subjects showed decreased D₂/D₃-receptor availability in the hypothalamus, mamillary bodies, thalamus and several thalamic nuclei, and increased glucose uptake in three lobules of the cerebellar vermis.

Conclusions: Dopaminergic system may be involved in modulation of grey matter metabolism and neurometabolic coupling in both healthy human brain and psychopathology. Hyperdopaminergic state in untreated schizophrenia may at least partly account for the corresponding decreases in grey matter metabolism.