TY - JOUR
T1 - Automated final lesion segmentation in posterior circulation acute ischemic stroke using deep learning
AU - Zoetmulder, Riaan
AU - Konduri, Praneeta R.
AU - Obdeijn, Iris V.
AU - Gavves, Efstratios
AU - Išgum, Ivana
AU - Majoie, Charles B.L.M.
AU - Dippel, Diederik W.J.
AU - Roos, Yvo B.W.E.M.
AU - Goyal, Mayank
AU - Mitchell, Peter J.
AU - Campbell, Bruce C.V.
AU - Lopes, Demetrius K.
AU - Reimann, Gernot
AU - Jovin, Tudor G.
AU - Saver, Jeffrey L.
AU - Muir, Keith W.
AU - White, Phil
AU - Bracard, Serge
AU - Chen, Bailiang
AU - Brown, Scott
AU - Schonewille, Wouter J.
AU - van der Hoeven, Erik
AU - Puetz, Volker
AU - Marquering, Henk A.
N1 - Acknowledgments: We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Quadro P6000 GPU used for this research.
Publisher Copyright:
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.
PY - 2021/9/4
Y1 - 2021/9/4
N2 - Final lesion volume (FLV) is a surrogate outcome measure in anterior circulation stroke (ACS). In posterior circulation stroke (PCS), this relation is plausibly understudied due to a lack of methods that automatically quantify FLV. The applicability of deep learning approaches to PCS is limited due to its lower incidence compared to ACS. We evaluated strategies to develop a convo-lutional neural network (CNN) for PCS lesion segmentation by using image data from both ACS and PCS patients. We included follow-up non-contrast computed tomography scans of 1018 patients with ACS and 107 patients with PCS. To assess whether an ACS lesion segmentation generalizes to PCS, a CNN was trained on ACS data (ACS-CNN). Second, to evaluate the performance of only including PCS patients, a CNN was trained on PCS data. Third, to evaluate the performance when combining the datasets, a CNN was trained on both datasets. Finally, to evaluate the performance of transfer learning, the ACS-CNN was fine-tuned using PCS patients. The transfer learning strategy outperformed the other strategies in volume agreement with an intra-class correlation of 0.88 (95% CI: 0.83–0.92) vs. 0.55 to 0.83 and a lesion detection rate of 87% vs. 41–77 for the other strategies. Hence, transfer learning improved the FLV quantification and detection rate of PCS lesions compared to the other strategies.
AB - Final lesion volume (FLV) is a surrogate outcome measure in anterior circulation stroke (ACS). In posterior circulation stroke (PCS), this relation is plausibly understudied due to a lack of methods that automatically quantify FLV. The applicability of deep learning approaches to PCS is limited due to its lower incidence compared to ACS. We evaluated strategies to develop a convo-lutional neural network (CNN) for PCS lesion segmentation by using image data from both ACS and PCS patients. We included follow-up non-contrast computed tomography scans of 1018 patients with ACS and 107 patients with PCS. To assess whether an ACS lesion segmentation generalizes to PCS, a CNN was trained on ACS data (ACS-CNN). Second, to evaluate the performance of only including PCS patients, a CNN was trained on PCS data. Third, to evaluate the performance when combining the datasets, a CNN was trained on both datasets. Finally, to evaluate the performance of transfer learning, the ACS-CNN was fine-tuned using PCS patients. The transfer learning strategy outperformed the other strategies in volume agreement with an intra-class correlation of 0.88 (95% CI: 0.83–0.92) vs. 0.55 to 0.83 and a lesion detection rate of 87% vs. 41–77 for the other strategies. Hence, transfer learning improved the FLV quantification and detection rate of PCS lesions compared to the other strategies.
UR - http://www.scopus.com/inward/record.url?scp=85114608266&partnerID=8YFLogxK
U2 - 10.3390/diagnostics11091621
DO - 10.3390/diagnostics11091621
M3 - Article
AN - SCOPUS:85114608266
VL - 11
JO - Diagnostics
JF - Diagnostics
SN - 2075-4418
IS - 9
M1 - 1621
ER -