| Title |
Analysis of Stroke Detection during the COVID-19 Pandemic Using Natural Language Processing of Radiology Reports
|
|---|---|
| Published in |
American Journal of Neuroradiology, December 2020
|
| DOI | 10.3174/ajnr.a6961 |
| Pubmed ID | |
| Authors |
M.D. Li, M. Lang, F. Deng, K. Chang, K. Buch, S. Rincon, W.A. Mehan, T.M. Leslie-Mazwi, J. Kalpathy-Cramer |
| Abstract |
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. 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. 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. 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. |
X Demographics
The data shown below were collected from the profiles of 11 X users who shared this research output. Click here to find out more about how the information was compiled.
Geographical breakdown
| Country | Count | As % |
|---|---|---|
| United States | 7 | 64% |
| Ireland | 1 | 9% |
| France | 1 | 9% |
| Brazil | 1 | 9% |
| Unknown | 1 | 9% |
Demographic breakdown
| Type | Count | As % |
|---|---|---|
| Members of the public | 6 | 55% |
| Science communicators (journalists, bloggers, editors) | 2 | 18% |
| Practitioners (doctors, other healthcare professionals) | 2 | 18% |
| Scientists | 1 | 9% |
Mendeley readers
The data shown below were compiled from readership statistics for 80 Mendeley readers of this research output. Click here to see the associated Mendeley record.
Geographical breakdown
| Country | Count | As % |
|---|---|---|
| Unknown | 80 | 100% |
Demographic breakdown
| Readers by professional status | Count | As % |
|---|---|---|
| Student > Bachelor | 10 | 13% |
| Researcher | 9 | 11% |
| Other | 8 | 10% |
| Student > Ph. D. Student | 6 | 8% |
| Student > Master | 5 | 6% |
| Other | 9 | 11% |
| Unknown | 33 | 41% |
| Readers by discipline | Count | As % |
|---|---|---|
| Medicine and Dentistry | 20 | 25% |
| Nursing and Health Professions | 7 | 9% |
| Neuroscience | 4 | 5% |
| Biochemistry, Genetics and Molecular Biology | 3 | 4% |
| Pharmacology, Toxicology and Pharmaceutical Science | 2 | 3% |
| Other | 8 | 10% |
| Unknown | 36 | 45% |
Attention Score in Context
This research output has an Altmetric Attention Score of 7. This is our high-level measure of the quality and quantity of online attention that it has received. This Attention Score, as well as the ranking and number of research outputs shown below, was calculated when the research output was last mentioned on 17 March 2021.
All research outputs
#4,779,792
of 23,269,984 outputs
Outputs from American Journal of Neuroradiology
#1,214
of 4,946 outputs
Outputs of similar age
#121,423
of 475,121 outputs
Outputs of similar age from American Journal of Neuroradiology
#39
of 92 outputs
Altmetric has tracked 23,269,984 research outputs across all sources so far. Compared to these this one has done well and is in the 79th percentile: it's in the top 25% of all research outputs ever tracked by Altmetric.
So far Altmetric has tracked 4,946 research outputs from this source. They typically receive a little more attention than average, with a mean Attention Score of 6.2. This one has done well, scoring higher than 75% of its peers.
Older research outputs will score higher simply because they've had more time to accumulate mentions. To account for age we can compare this Altmetric Attention Score to the 475,121 tracked outputs that were published within six weeks on either side of this one in any source. This one has gotten more attention than average, scoring higher than 74% of its contemporaries.
We're also able to compare this research output to 92 others from the same source and published within six weeks on either side of this one. This one has gotten more attention than average, scoring higher than 57% of its contemporaries.