Predicting final ischemic stroke lesions from initial diffusion-weighted images using a deep neural network

For prognosis of stroke, measurement of the diffusion-perfusion mismatch is a common practice for estimating tissue at risk of infarction in the absence of timely reperfusion. However, perfusion-weighted imaging (PWI) adds time and expense to the acute stroke imaging workup. We explored whether a deep convolutional neural network (DCNN) model trained with diffusion-weighted imaging obtained at admission could predict final infarct volume and location in acute stroke patients. In 445 patients, we trained and validated an attention-gated (AG) DCNN to predict final infarcts as delineated on follow-up studies obtained 3 to 7 days after stroke. The input channels consisted of MR diffusion-weighted imaging (DWI), apparent diffusion coefficients (ADC) maps, and thresholded ADC maps with values less than 620 × 10−6 mm2/s, while the output was a voxel-by-voxel probability map of tissue infarction. We evaluated performance of the model using the area under the receiver-operator characteristic curve (AUC), the Dice similarity coefficient (DSC), absolute lesion volume error, and the concordance correlation coefficient (ρc) of the predicted and true infarct volumes. The model obtained a median AUC of 0.91 (IQR: 0.84–0.96). After thresholding at an infarction probability of 0.5, the median sensitivity and specificity were 0.60 (IQR: 0.16–0.84) and 0.97 (IQR: 0.93–0.99), respectively, while the median DSC and absolute volume error were 0.50 (IQR: 0.17–0.66) and 27 ml (IQR: 7–60 ml), respectively. The model’s predicted lesion volumes showed high correlation with ground truth volumes (ρc = 0.73, p < 0.01). An AG-DCNN using diffusion information alone upon admission was able to predict infarct volumes at 3–7 days after stroke onset with comparable accuracy to models that consider both DWI and PWI. This may enable treatment decisions to be made with shorter stroke imaging protocols.