Department of Radiology (Y.L., Y.Y., J.O., B.J., S.O., G.Z.).
Department of Electrical Engineering (J.O.), Stanford University, CA.
Stroke. 2023 Sep;54(9):2316-2327. doi: 10.1161/STROKEAHA.123.044072. Epub 2023 Jul 24.
Predicting long-term clinical outcome based on the early acute ischemic stroke information is valuable for prognostication, resource management, clinical trials, and patient expectations. Current methods require subjective decisions about which imaging features to assess and may require time-consuming postprocessing. This study's goal was to predict ordinal 90-day modified Rankin Scale (mRS) score in acute ischemic stroke patients by fusing a Deep Learning model of diffusion-weighted imaging images and clinical information from the acute period.
A total of 640 acute ischemic stroke patients who underwent magnetic resonance imaging within 1 to 7 days poststroke and had 90-day mRS follow-up data were randomly divided into 70% (n=448) for model training, 15% (n=96) for validation, and 15% (n=96) for internal testing. Additionally, external testing on a cohort from Lausanne University Hospital (n=280) was performed to further evaluate model generalization. Accuracy for ordinal mRS, accuracy within ±1 mRS category, mean absolute prediction error, and determination of unfavorable outcome (mRS score >2) were evaluated for clinical only, imaging only, and 2 fused clinical-imaging models.
The fused models demonstrated superior performance in predicting ordinal mRS score and unfavorable outcome in both internal and external test cohorts when compared with the clinical and imaging models. For the internal test cohort, the top fused model had the highest area under the curve of 0.92 for unfavorable outcome prediction and the lowest mean absolute error (0.96 [95% CI, 0.77-1.16]), with the highest proportion of mRS score predictions within ±1 category (79% [95% CI, 71%-88%]). On the external Lausanne University Hospital cohort, the best fused model had an area under the curve of 0.90 for unfavorable outcome prediction and outperformed other models with an mean absolute error of 0.90 (95% CI, 0.79-1.01), and the highest percentage of mRS score predictions within ±1 category (83% [95% CI, 78%-87%]).
A Deep Learning-based imaging model fused with clinical variables can be used to predict 90-day stroke outcome with reduced subjectivity and user burden.
基于早期急性缺血性卒中信息预测长期临床结局对于预后评估、资源管理、临床试验和患者预期具有重要价值。目前的方法需要主观决定评估哪些成像特征，并且可能需要耗时的后处理。本研究的目的是通过融合弥散加权成像图像的深度学习模型和急性期的临床信息来预测急性缺血性卒中患者的 90 天改良 Rankin 量表（mRS）评分。
共纳入 640 例发病后 1 至 7 天内行磁共振成像检查且有 90 天 mRS 随访数据的急性缺血性卒中患者，将其随机分为 70%（n=448）用于模型训练、15%（n=96）用于验证、15%（n=96）用于内部测试。此外，还对洛桑大学医院的队列（n=280）进行了外部测试，以进一步评估模型的泛化能力。评估了仅临床、仅成像和 2 种融合临床成像模型对 ordinal mRS、mRS 类别内准确率、平均绝对预测误差和不良结局（mRS 评分>2）的预测准确性。
与临床和成像模型相比，融合模型在内部和外部测试队列中均能更好地预测 ordinal mRS 评分和不良结局。对于内部测试队列，最佳融合模型对不良结局预测的曲线下面积最高为 0.92，平均绝对误差最低（0.96 [95%CI，0.77-1.16]），mRS 评分类别内预测比例最高（79% [95%CI，71%-88%]）。在外部洛桑大学医院队列中，最佳融合模型对不良结局预测的曲线下面积为 0.90，平均绝对误差优于其他模型（0.90 [95%CI，0.79-1.01]），mRS 评分类别内预测比例最高（83% [95%CI，78%-87%]）。
融合临床变量的基于深度学习的成像模型可用于预测 90 天卒中结局，可降低主观性和用户负担。
