Predicting DWI-FLAIR mismatch on NCCT: the role of artificial intelligence in hyperacute decision ...

@article{2367132,
   author = {Kim, Beom Joon and Zhu, Kairan and Qiu, Wu and Singh, Nishita and McDonough, Rosalie and Cimflová, Petra and Bala, Fouzi and Kim, Jongwook and Kim, Yong Soo and Bae, HeeandJoon and Menon, Bijoy K},
   article_location = {Lausanne},
   article_number = {June},
   doi = {http://dx.doi.org/10.3389/fneur.2023.1201223},
   keywords = {artificial intelligence; DWI; FLAIR; DWI-FLAIR mismatch; non-contrast computed tomography},
   language = {eng},
   issn = {1664-2295},
   journal = {Frontiers in Neurology},
   title = {Predicting DWI-FLAIR mismatch on NCCT: the role of artificial intelligence in hyperacute decision making},
   url = {https://www.frontiersin.org/articles/10.3389/fneur.2023.1201223/full},
   volume = {14},
   year = {2023}
}

TY  - JOUR
ID  - 2367132
AU  - Kim, Beom Joon - Zhu, Kairan - Qiu, Wu - Singh, Nishita - McDonough, Rosalie - Cimflová, Petra - Bala, Fouzi - Kim, Jongwook - Kim, Yong Soo - Bae, Hee-Joon - Menon, Bijoy K
PY  - 2023
TI  - Predicting DWI-FLAIR mismatch on NCCT: the role of artificial intelligence in hyperacute decision making
JF  - Frontiers in Neurology
VL  - 14
IS  - June
SP  - 2023
EP  - 2023
PB  - Frontiers
SN  - 16642295
KW  - artificial intelligence
KW  - DWI
KW  - FLAIR
KW  - DWI-FLAIR mismatch
KW  - non-contrast computed tomography
UR  - https://www.frontiersin.org/articles/10.3389/fneur.2023.1201223/full
N2  - BackgroundThe presence of diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) mismatch was used to determine eligibility for intravenous thrombolysis in clinical trials. However, due to the restricted availability of MRI and the ambiguity of image assessment, it is not widely implemented in clinical practice. MethodsA total of 222 acute ischemic stroke patients underwent non-contrast computed tomography (NCCT), DWI, and FLAIR within 1 h of one another. Human experts manually segmented ischemic lesions on DWI and FLAIR images and independently graded the presence of DWI-FLAIR mismatch. Deep learning (DL) models based on the nnU-net architecture were developed to predict ischemic lesions visible on DWI and FLAIR images using NCCT images. Inexperienced neurologists evaluated the DWI-FLAIR mismatch on NCCT images without and with the model's results. ResultsThe mean age of included subjects was 71.8 & PLUSMN; 12.8 years, 123 (55%) were male, and the baseline NIHSS score was a median of 11 [IQR, 6-18]. All images were taken in the following order: NCCT - DWI - FLAIR, starting after a median of 139 [81-326] min after the time of the last known well. Intravenous thrombolysis was administered in 120 patients (54%) after NCCT. The DL model's prediction on NCCT images revealed a Dice coefficient and volume correlation of 39.1% and 0.76 for DWI lesions and 18.9% and 0.61 for FLAIR lesions. In the subgroup with 15 mL or greater lesion volume, the evaluation of DWI-FLAIR mismatch from NCCT by inexperienced neurologists improved in accuracy (from 0.537 to 0.610) and AUC-ROC (from 0.493 to 0.613). ConclusionThe DWI-FLAIR mismatch may be reckoned using NCCT images through advanced artificial intelligence techniques.
ER  - 

KIM, Beom Joon, Kairan ZHU, Wu QIU, Nishita SINGH, Rosalie MCDONOUGH, Petra CIMFLOVÁ, Fouzi BALA, Jongwook KIM, Yong Soo KIM, Hee-Joon BAE and Bijoy K MENON. Predicting DWI-FLAIR mismatch on NCCT: the role of artificial intelligence in hyperacute decision making. \textit{Frontiers in Neurology}. Lausanne: Frontiers, 2023, vol.~14, June, p.~2023-2032. ISSN~1664-2295. Available from: https://dx.doi.org/10.3389/fneur.2023.1201223.