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

Beom Joon Kim; Kairan Zhu; Wu Qiu; Nishita Singh; Rosalie McDonough; Petra Cimflova; Fouzi Bala; Jongwook Kim; Yong Soo Kim; Hee-Joon Bae; Bijoy K Menon

doi:10.3389/fneur.2023.1201223

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

Front Neurol. 2023 Jun 12:14:1201223. doi: 10.3389/fneur.2023.1201223. eCollection 2023.

Authors

Beom Joon Kim^{1

2}, Kairan Zhu³, Wu Qiu⁴, Nishita Singh^{5

6}, Rosalie McDonough⁵, Petra Cimflova^{5

7}, Fouzi Bala^{5

8}, Jongwook Kim¹, Yong Soo Kim⁹, Hee-Joon Bae^{1

10}, Bijoy K Menon⁵

Affiliations

¹ Department of Neurology, Seoul National University Bundang Hospital, Seongnam-si, Republic of Korea.
² Gyeonggi Regional Cerebrovascular Center, Seoul National University Bundang Hospital, Seongnam-si, Republic of Korea.
³ College of Electronic Engineering, Xi'an Shiyou University, Xi'an, Shaanxi, China.
⁴ School of Life Science and Technology, Huazhong University of Science and Technology, Wuhan, Hubei, China.
⁵ Department of Clinical Neurosciences and Diagnostic Imaging, University of Calgary Cumming School of Medicine, Calgary, AB, Canada.
⁶ Neurology Division, Department of Internal Medicine, Max Rady College of Medicine, University of Manitoba, Winnipeg, MB, Canada.
⁷ Department of Medical Imaging, St Anne's University Hospital Brno and Faculty of Medicine, Masaryk University, Brno, Czechia.
⁸ Diagnostic and Interventional Neuroradiology Department, University Hospital of Tours, Tours, France.
⁹ Department of Neurology, Nowon Eulji Medical Center, Eulji University, Seoul, Republic of Korea.
¹⁰ Department of Neurology, College of Medicine, Seoul National University, Seoul, Republic of Korea.

Abstract

Background: The 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.

Methods: A 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.

Results: The mean age of included subjects was 71.8 ± 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).

Conclusion: The DWI-FLAIR mismatch may be reckoned using NCCT images through advanced artificial intelligence techniques.

Keywords: DWI; DWI-FLAIR mismatch; FLAIR; artificial intelligence; non-contrast computed tomography.