arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2412.17840eess.SP
收藏

基于约束模型的多模态学习在改善室性心律失常预测中的应用

Constraint-Based Model in Multimodal Learning to Improve Ventricular Arrhythmia Prediction

Evariste Njomgue Fotso, Buntheng Ly, Hubert Cochet, Maxime Sermesant

AI总结:

本研究针对室性心律失常预测,提出基于先验知识约束医学图像学习过程的Sequential Fusion多模态融合方法,利用图卷积网络提取3D心肌厚度图特征,实验证明其性能优于传统Early Fusion和单模态模型。

摘要:

心脏疾病评估依赖于多种诊断模态:心电图(ECG)用于诊断心律异常,以及磁共振成像(MRI)、计算机断层扫描(CT)和超声心动图等成像模态用于检测结构异常的迹象。这些模态各自带来互补信息,以更好地诊断心功能不全。然而,使用来自多种模态的数据训练机器学习(ML)模型是一项具有挑战性的任务,因为它增加了维度空间,同时保持样本数量不变。事实上,随着输入空间维度的增加,实现准确泛化所需的数据量呈指数级增长。在本研究中,针对基于临床和CT成像特征组合的室性心律失常(VA)预测应用,我们解决了这一问题,其中我们基于从临床数据中获得的先验知识,对医学图像(CT)的学习过程施加了约束。VA分类器接收从左心室的3D心肌厚度图(TM)中提取的特征作为输入。该TM由我们的处理流程从成像输入生成,并使用图卷积网络作为3D TM的特征提取器。我们引入了一种新颖的Sequential Fusion方法,并将其性能与传统的Early Fusion技术及单模态模型进行了比较。交叉验证结果表明,Sequential Fusion模型取得了最高的平均得分,灵敏度(Sensitivity)为80.7% ± 4.4,F1分数为73.1% ± 6.0,优于灵敏度(Sensitivity)为65.0% ± 8.9、F1分数为63.1% ± 6.3的Early Fusion模型。两种融合模型的得分均优于单模态模型,其中临床数据模态的平均灵敏度(Sensitivity)和F1分数分别为62.8% ± 10.1和52.1% ± 6.5,医学图像模态的平均灵敏度(Sensitivity)和F1分数分别为62.9% ± 6.3和60.7% ± 5.3。

英文摘要:

Cardiac disease evaluation depends on multiple diagnostic modalities: electrocardiogram (ECG) to diagnose abnormal heart rhythms, and imaging modalities such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT) and echocardiography to detect signs of structural abnormalities. Each of these modalities brings complementary information for a better diagnosis of cardiac dysfunction. However, training a machine learning (ML) model with data from multiple modalities is a challenging task, as it increases the dimension space, while keeping constant the number of samples. In fact, as the dimension of the input space increases, the volume of data required for accurate generalisation grows exponentially. In this work, we address this issue, for the application of Ventricular Arrhythmia (VA) prediction, based on the combined clinical and CT imaging features, where we constrained the learning process on medical images (CT) based on the prior knowledge acquired from clinical data. The VA classifier is fed with features extracted from a 3D myocardium thickness map (TM) of the left ventricle. The TM is generated by our pipeline from the imaging input and a Graph Convolutional Network is used as the feature extractor of the 3D TM. We introduce a novel Sequential Fusion method and evaluate its performance against traditional Early Fusion techniques and single-modality models. The crossvalidation results show that the Sequential Fusion model achieved the highest average scores of 80.7% $\pm$ 4.4 Sensitivity and 73.1% $\pm$ 6.0 F1 score, outperforming the Early Fusion model at 65.0% $\pm$ 8.9 Sensitivity and 63.1% $\pm$ 6.3 F1 score. Both fusion models achieved better scores than the single-modality models, where the average Sensitivity and F1 score are 62.8% $\pm$ 10.1; 52.1% $\pm$ 6.5 for the clinical data modality and 62.9% $\pm$ 6.3; 60.7% $\pm$ 5.3 for the medical images modality.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑