Objective: To investigate the performance of a commercially available artificial intelligence (AI) algorithm for the detection of pulmonary embolism (PE) on contrast-enhanced computed tomography (CT) scans in patients hospitalized for coronavirus disease 2019 (COVID-19).
Patients And Methods: Retrospective analysis was performed of all contrast-enhanced chest CT scans of patients admitted for COVID-19 between March 1, 2020 and December 31, 2021. Based on the original radiology reports, all PE-positive examinations were included (n=527). Using a reversed-flow single-gate diagnostic accuracy case-control model, a randomly selected cohort of PE-negative examinations (n=977) was included. Pulmonary parenchymal disease severity was assessed for all the included studies using a semiquantitative system, the total severity score. All included CT scans were sent for interpretation by the commercially available AI algorithm, Aidoc. Discrepancies between AI and original radiology reports were resolved by 3 blinded radiologists, who rendered a final determination of indeterminate, positive, or negative.
Results: A total of 78 studies were found to be discrepant, of which 13 (16.6%) were deemed indeterminate by readers and were excluded. The sensitivity and specificity of AI were 93.2% (95% CI, 90.6%-95.2%) and 99.6% (95% CI, 98.9%-99.9%), respectively. The accuracy of AI for all total severity score groups (mild, moderate, and severe) was high (98.4%, 96.7%, and 97.2%, respectively). Artificial intelligence was more accurate in PE detection on CT pulmonary angiography scans than on contrast-enhanced CT scans (<.001), with an optimal Hounsfield unit of 362 (=.048).
Conclusion: The AI algorithm demonstrated high sensitivity, specificity, and accuracy for PE on contrast-enhanced CT scans in patients with COVID-19 regardless of parenchymal disease. Accuracy was significantly affected by the mean attenuation of the pulmonary vasculature. How this affects the legitimacy of the binary outcomes reported by AI is not yet known.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9995315 | PMC |
| http://dx.doi.org/10.1016/j.mayocpiqo.2023.03.001 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Brain Iron Deposition Alterations in Type 2 Diabetes Mellitus Patients With Mild Cognitive Impairment Based on Quantitative Susceptibility Mapping.
J Diabetes
January 2025
Department of Radiology, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing City, Jiangsu Province, China.
Background: Iron is one of the most important elements in brain that may has a direct impact on the stability of central nervous system. The current study devoted to explore the alterations of iron distribution across the whole brain in type 2 diabetes mellitus (T2DM) patients with mild cognitive impairment (MCI).
Methods: The quantitative susceptibility mapping (QSM) technique was used to quantify the intracranial iron content of 74 T2DM patients with MCI and 86 T2DM patients with normal cognition (NC).
Diagnostic impact of DNA methylation classification in adult and pediatric CNS tumors.
Sci Rep
January 2025
DIAPath, Center for Microscopy and Molecular Imaging (CMMI), Université Libre de Bruxelles (ULB), 6041, Gosselies, Belgium.
Over the past decade, neuropathological diagnosis has undergone significant changes, integrating morphological features with molecular biomarkers. The molecular era has successfully refined neuropathological diagnostic accuracy; however, a substantial number of CNS tumor diagnoses remain challenging, particularly in children. DNA methylation classification has emerged as a powerful machine learning approach for clinical decision-making in CNS tumors.
View Article and Find Full Text PDFChallenges and opportunities of vaccination during pregnancy: perspectives of 20 healthcare professionals.
J Public Health Policy
January 2025
George's School of Health and Medical Sciences, Population Health Research Institute, City St George's, University of London, London, UK.
Vaccination during pregnancy is crucial due to increased maternal vulnerability to infectious diseases. However, uptake of recommended vaccines (influenza, pertussis, COVID-19) remains suboptimal, particularly among disadvantaged groups. This qualitative study explored healthcare professionals' (HCPs) perspectives, selected purposively, on factors influencing maternal vaccination in London.
View Article and Find Full Text PDFNeural processing of naturalistic audiovisual events in space and time.
Commun Biol
January 2025
Western Institute for Neuroscience, Western University, London, ON, Canada.
Our brain seamlessly integrates distinct sensory information to form a coherent percept. However, when real-world audiovisual events are perceived, the specific brain regions and timings for processing different levels of information remain less investigated. To address that, we curated naturalistic videos and recorded functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) data when participants viewed videos with accompanying sounds.
View Article and Find Full Text PDFMedical image segmentation based on frequency domain decomposition SVD linear attention.
Sci Rep
January 2025
Affiliated Hospital 6 of Nantong University, Yancheng Third People's Hospital, Yancheng, 224001, Jiangsu, China.
Convolutional Neural Networks (CNNs) have achieved remarkable segmentation accuracy in medical image segmentation tasks. However, the Vision Transformer (ViT) model, with its capability of extracting global information, offers a significant advantage in contextual information compared to the limited receptive field of convolutional kernels in CNNs. Despite this, ViT models struggle to fully detect and extract high-frequency signals, such as textures and boundaries, in medical images.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!