Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection
BackgroundNoncontrast CT (NCCT) is used to evaluate for intracerebral hemorrhage (ICH) and ischemia in acute ischemic stroke (AIS). Large vessel occlusions (LVOs) are a major cause of AIS, but challenging to detect on NCCT.AimsThe purpose of this study is to evaluate an AI software called RAPID NCCT...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-11-01
|
| Series: | Frontiers in Neurology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fneur.2023.1324088/full |
| _version_ | 1797396925301915648 |
|---|---|
| author | Vivek Yedavalli Jeremy Josef Heit Seena Dehkharghani Hafez Haerian John Mcmenamy Justin Honce Vincent M. Timpone Christopher Harnain Andrew Kesselman Anthony Filly Adam Beardsley Brian Sakamoto Chris Song James Montuori Benjamin Navot Francisco Villar Mena Dan-Victor Giurgiutiu Felipe Kitamura Fabricio Oliveira Lima Henrique Silva Francisco Jose Mont’Alverne Gregory Albers |
| author_facet | Vivek Yedavalli Jeremy Josef Heit Seena Dehkharghani Hafez Haerian John Mcmenamy Justin Honce Vincent M. Timpone Christopher Harnain Andrew Kesselman Anthony Filly Adam Beardsley Brian Sakamoto Chris Song James Montuori Benjamin Navot Francisco Villar Mena Dan-Victor Giurgiutiu Felipe Kitamura Fabricio Oliveira Lima Henrique Silva Francisco Jose Mont’Alverne Gregory Albers |
| author_sort | Vivek Yedavalli |
| collection | DOAJ |
| description | BackgroundNoncontrast CT (NCCT) is used to evaluate for intracerebral hemorrhage (ICH) and ischemia in acute ischemic stroke (AIS). Large vessel occlusions (LVOs) are a major cause of AIS, but challenging to detect on NCCT.AimsThe purpose of this study is to evaluate an AI software called RAPID NCCT Stroke (RAPID, iSchemaView, Menlo Park, CA) for ICH and LVO detection compared to expert readers.MethodsIn this IRB approved retrospective, multicenter study, stand-alone performance of the software was assessed based on the consensus of 3 neuroradiologists and sensitivity and specificity were determined. The platform’s performance was then compared to interpretation by readers comprised of eight general radiologists (GR) and three neuroradiologists (NR) in detecting ICH and hyperdense vessel sign (HVS) indicating LVO.ResultsA total of 244 cases were included. Of the 244, 115 were LVOs and 26 were ICHs. One hundred three cases did not have LVO nor ICH. Stand-alone performance of the software demonstrated sensitivities and specificities of 96.2 and 99.5% for ICH and 63.5 and 95.1% for LVO detection. Compared to all 11 readers and eight GR readers only respectively, the software demonstrated superiority, achieving significantly higher sensitivities (63.5% versus 43.6%, p < 0.0001 and 63.5% versus 40.9%, p = 0.001).ConclusionThe RAPID NCCT Stroke platform demonstrates superior performance to radiologists for detecting LVO from a NCCT. Use of this software platform could lead to earlier LVO detection and expedited transfer of these patients to a thrombectomy capable center. |
| first_indexed | 2024-03-09T00:58:23Z |
| format | Article |
| id | doaj.art-4d3d53b88e13448faf4da57338aaf0a2 |
| institution | Directory Open Access Journal |
| issn | 1664-2295 |
| language | English |
| last_indexed | 2024-03-09T00:58:23Z |
| publishDate | 2023-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurology |
| spelling | doaj.art-4d3d53b88e13448faf4da57338aaf0a22023-12-11T16:23:52ZengFrontiers Media S.A.Frontiers in Neurology1664-22952023-11-011410.3389/fneur.2023.13240881324088Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detectionVivek Yedavalli0Jeremy Josef Heit1Seena Dehkharghani2Hafez Haerian3John Mcmenamy4Justin Honce5Vincent M. Timpone6Christopher Harnain7Andrew Kesselman8Anthony Filly9Adam Beardsley10Brian Sakamoto11Chris Song12James Montuori13Benjamin Navot14Francisco Villar Mena15Dan-Victor Giurgiutiu16Felipe Kitamura17Fabricio Oliveira Lima18Henrique Silva19Francisco Jose Mont’Alverne20Gregory Albers21The Johns Hopkins Hospital, Johns Hopkins Medicine, Baltimore, MD, United StatesStanford Healthcare, Standford, CA, United StatesDepartment of Radiology, New York University, New York, NY, United StatesLifeBridge Health, Baltimore, MD, United StatesDepartment of Radiology, New York University, New York, NY, United StatesDepartment of Radiology, University of Colorado, Denver, CO, United StatesMayo Clinic, Rochester, MN, United StatesWeill Cornell Medicine, Cornell University, New York, NY, United StatesDepartment of Radiology, Stanford University, Standford, CA, United StatesStanford Healthcare, Standford, CA, United StatesDepartment of Radiology, University of Virginia Hospital, Charlottesville, VA, United States0Dameron Hospital, Stockton, CA, United StatesWeill Cornell Medicine, Cornell University, New York, NY, United States1Capital Health, Trenton, NJ, United States2Columbia College, Columbia University, New York, NY, United States3Las Condes Clinic, Santiago, Chile4Department of Radiology, Augusta University, Augusta, GA, United States5Department of Radiology, Universidade Federal de São Paulo, Dasa, Brazil6Hospital Geral de Fortaleza, Fortaleza, Brazil6Hospital Geral de Fortaleza, Fortaleza, Brazil7Santa Casa de Misericórdia de Porto Alegre, New York, NY, BrazilStanford Healthcare, Standford, CA, United StatesBackgroundNoncontrast CT (NCCT) is used to evaluate for intracerebral hemorrhage (ICH) and ischemia in acute ischemic stroke (AIS). Large vessel occlusions (LVOs) are a major cause of AIS, but challenging to detect on NCCT.AimsThe purpose of this study is to evaluate an AI software called RAPID NCCT Stroke (RAPID, iSchemaView, Menlo Park, CA) for ICH and LVO detection compared to expert readers.MethodsIn this IRB approved retrospective, multicenter study, stand-alone performance of the software was assessed based on the consensus of 3 neuroradiologists and sensitivity and specificity were determined. The platform’s performance was then compared to interpretation by readers comprised of eight general radiologists (GR) and three neuroradiologists (NR) in detecting ICH and hyperdense vessel sign (HVS) indicating LVO.ResultsA total of 244 cases were included. Of the 244, 115 were LVOs and 26 were ICHs. One hundred three cases did not have LVO nor ICH. Stand-alone performance of the software demonstrated sensitivities and specificities of 96.2 and 99.5% for ICH and 63.5 and 95.1% for LVO detection. Compared to all 11 readers and eight GR readers only respectively, the software demonstrated superiority, achieving significantly higher sensitivities (63.5% versus 43.6%, p < 0.0001 and 63.5% versus 40.9%, p = 0.001).ConclusionThe RAPID NCCT Stroke platform demonstrates superior performance to radiologists for detecting LVO from a NCCT. Use of this software platform could lead to earlier LVO detection and expedited transfer of these patients to a thrombectomy capable center.https://www.frontiersin.org/articles/10.3389/fneur.2023.1324088/fullstrokeacuteischemic strokenoncontrast CTlarge vessel occlusion |
| spellingShingle | Vivek Yedavalli Jeremy Josef Heit Seena Dehkharghani Hafez Haerian John Mcmenamy Justin Honce Vincent M. Timpone Christopher Harnain Andrew Kesselman Anthony Filly Adam Beardsley Brian Sakamoto Chris Song James Montuori Benjamin Navot Francisco Villar Mena Dan-Victor Giurgiutiu Felipe Kitamura Fabricio Oliveira Lima Henrique Silva Francisco Jose Mont’Alverne Gregory Albers Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection Frontiers in Neurology stroke acute ischemic stroke noncontrast CT large vessel occlusion |
| title | Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection |
| title_full | Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection |
| title_fullStr | Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection |
| title_full_unstemmed | Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection |
| title_short | Performance of RAPID noncontrast CT stroke platform in large vessel occlusion and intracranial hemorrhage detection |
| title_sort | performance of rapid noncontrast ct stroke platform in large vessel occlusion and intracranial hemorrhage detection |
| topic | stroke acute ischemic stroke noncontrast CT large vessel occlusion |
| url | https://www.frontiersin.org/articles/10.3389/fneur.2023.1324088/full |
| work_keys_str_mv | AT vivekyedavalli performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT jeremyjosefheit performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT seenadehkharghani performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT hafezhaerian performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT johnmcmenamy performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT justinhonce performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT vincentmtimpone performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT christopherharnain performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT andrewkesselman performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT anthonyfilly performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT adambeardsley performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT briansakamoto performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT chrissong performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT jamesmontuori performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT benjaminnavot performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT franciscovillarmena performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT danvictorgiurgiutiu performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT felipekitamura performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT fabriciooliveiralima performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT henriquesilva performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT franciscojosemontalverne performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection AT gregoryalbers performanceofrapidnoncontrastctstrokeplatforminlargevesselocclusionandintracranialhemorrhagedetection |