基于MRI的血管周围间隙扩大负担自动分级的多任务学习
Multi-task learning for the automatic grading of enlarged perivascular space burden using MRI
另 4 家 · 查看机构详情
- University of Edinburgh（爱丁堡大学）
- Canon Medical Research Europe（佳能医疗研究欧洲）
- UK Dementia Research Institute（英国痴呆症研究所）
- Chongqing General Hospital（重庆市人民医院）
- Chongqing University（重庆大学）
- University of California, San Francisco（加利福尼亚大学旧金山分校）
- University of Aberdeen（阿伯丁大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本研究提出多任务学习模型，同时进行PVS分割与评分，在MRI影像上自动分级血管周围间隙扩大负担，平均精度达64.08%，优于单任务CNN和逻辑回归方法。
中文摘要 AI 辅助
脑磁共振成像（MRI）中可见的血管周围间隙扩大（PVS）日益被认为与脑健康状况不佳有关。PVS是直径小于3毫米的细长结构，且数量可能众多。为了反映PVS的发生率，放射科医生按照临床分级量表对其负担进行视觉评分——这一任务若能自动化将有助于加速分析并克服观察者间差异的影响。我们开发并评估了训练机器学习模型的方法，以利用Potters/Wardlaw量表对基底节（BG）和半卵圆中心（CSO）的PVS发生率进行评分。我们工作的新颖之处在于，在训练过程中使用了不完美的、半自动生成的“银标准”PVS分割掩膜，以及PVS放射学评分。我们比较评估了一个条件卷积神经网络（CNN），它接受PVS掩膜作为额外输入通道；一个多任务CNN，它同时执行PVS分割和评分；以及一个逻辑回归模型，它利用从PVS掩膜提取的特征来预测PVS评分。多任务学习是最有效的方法，其平均精度均值达到64.08%，而条件CNN为60.22%，仅训练用于预测PVS评分的基线CNN为52.11%，逻辑回归模型为49.32%。多任务模型展示了定位单个PVS的能力，这是其他CNN所不具备的，并且其行为在概率上是合理的，对本质上较难的类别以较低的置信度进行预测。年龄、性别、高血压状态、白质高信号体积和缺血性卒中病变状态被证明与多任务模型的PVS评分预测以及真实标签以相似的方式相关联。
英文摘要
Enlarged perivascular spaces (PVS) visible in brain magnetic resonance imaging (MRI) are increasingly thought to be linked to poor brain health. PVS are elongated structures of less than 3 mm in diameter and can be numerous. To reflect the incidence of PVS, radiologists visually score their burden following a clinical grading scale - a task that would benefit from automation to accelerate analyses and overcome the influence of inter-observer differences. We developed and evaluated methods for training machine learning models to score PVS incidence in the basal ganglia (BG) and centrum semiovale (CSO) leveraging the Potters/Wardlaw scale. The novelty in our work lies in the use of imperfect, semi-automatically generated "silver-standard" PVS segmentation masks during training, in addition to PVS radiological scores. We comparatively evaluated a conditional convolutional neural network (CNN) which accepts PVS masks as an extra input channel, a multi-task CNN which performs both PVS segmentation and scoring, and a logistic regression model which utilises features derived from PVS masks to predict PVS scores. Multi-task learning was the most effective method, achieving a mean average precision of 64.08% compared to 60.22% for the conditional CNN, 52.11% for a baseline CNN trained only to predict PVS scores, and 49.32% for the logistic regression model. The multi-task model showed an ability to localise individual PVS not shown by the other CNNs, and behaved in a probabilistically sensible way, predicting with lower confidence on inherently harder classes. Age, sex, hypertension status, white matter hyperintensity volume, and ischaemic stroke lesion status were shown to be associated with the multi-task model's PVS score predictions and the ground truth in a similar way.
