Litcius/Paper detail

Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke

Yannan Yu, Yuan Xie, Thoralf Thamm, Enhao Gong, Jiahong Ouyang, Sören Christensen, Michael P. Marks, Maarten G. Lansberg, Gregory W. Albers, Greg Zaharchuk

2021American Journal of Neuroradiology34 citationsDOIOpen Access PDF

Abstract

BACKGROUND AND PURPOSE: In acute stroke patients with large vessel occlusions, it would be helpful to be able to predict the difference in the size and location of the final infarct based on the outcome of reperfusion therapy. Our aim was to demonstrate the value of deep learning-based tissue at risk and ischemic core estimation. We trained deep learning models using a baseline MR image in 3 multicenter trials. MATERIALS AND METHODS: Patients with acute ischemic stroke from 3 multicenter trials were identified and grouped into minimal (≤20%), partial (20%-80%), and major (≥80%) reperfusion status based on 4- to 24-hour follow-up MR imaging if available or into unknown status if not. Attention-gated convolutional neural networks were trained with admission imaging as input and the final infarct as ground truth. We explored 3 approaches: 1) separate: train 2 independent models with patients with minimal and major reperfusion; 2) pretraining: develop a single model using patients with partial and unknown reperfusion, then fine-tune it to create 2 separate models for minimal and major reperfusion; and 3) thresholding: use the current clinical method relying on apparent diffusion coefficient and time-to-maximum of the residue function maps. Models were evaluated using area under the curve, the Dice score coefficient, and lesion volume difference. RESULTS: < .001). CONCLUSIONS: Deep learning models with fine-tuning lead to better performance for predicting tissue at risk and ischemic core, outperforming conventional thresholding methods.

Topics & Concepts

MedicineStroke (engine)Ischemic strokeCore (optical fiber)EstimationArtificial intelligenceAcute strokeDeep learningCardiologyInternal medicineIschemiaTissue plasminogen activatorComputer scienceEngineeringMechanical engineeringManagementEconomicsTelecommunicationsAcute Ischemic Stroke ManagementGenerative Adversarial Networks and Image SynthesisAdvanced MRI Techniques and Applications
Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke | Litcius