Predicting recovery following stroke: deep learning, multimodal data and feature selection using explainable AI

📄 arXiv: 2310.19174v1 📥 PDF

作者: Adam White, Margarita Saranti, Artur d'Avila Garcez, Thomas M. H. Hope, Cathy J. Price, Howard Bowman

分类: cs.AI, cs.CY, cs.LG

发布日期: 2023-10-29

💡 一句话要点

提出深度学习方法以提高中风后恢复预测的准确性

🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 中风恢复预测 深度学习 多模态数据 卷积神经网络 特征选择 神经影像分析 分类准确性

📋 核心要点

  1. 现有方法面临神经影像数据维度高、数据集小及多模态数据结合困难等挑战。
  2. 论文提出通过CNN结合MRI图像和表格数据的特征选择来提高中风后恢复预测的准确性。
  3. 实验结果显示，使用8个感兴趣区域和其他特征的2D残差神经网络达到了0.854的分类准确率，显著提升了基线模型的性能。

📝 摘要（中文）

机器学习在自动预测中风后症状及其康复反应方面具有巨大潜力。然而，神经影像数据的高维性、可用数据集的相对较小以及如何有效结合神经影像与表格数据（如人口统计信息和临床特征）是主要挑战。本文评估了基于两种策略的几种解决方案：使用总结MRI扫描的2D图像和选择关键特征以提高分类准确性。我们引入了一种新颖的方法，通过卷积神经网络（CNN）训练结合从MRI中提取的感兴趣区域与表格数据的符号表示的图像。实验结果表明，结合影像和表格数据可以实现高分类准确性，尽管数据集在机器学习中较小。

🔬 方法详解

问题定义：本文旨在解决中风后恢复预测的准确性问题，现有方法在处理高维神经影像数据和小数据集时效果不佳。

核心思路：通过结合MRI图像的感兴趣区域与表格数据，利用卷积神经网络（CNN）进行特征选择和分类，以提高预测准确性。

技术框架：整体架构包括数据预处理、特征提取、CNN模型训练和分类评估等主要模块。首先，提取MRI图像的关键区域，然后与临床特征结合，输入到CNN中进行训练。

关键创新：引入了结合MRI图像和表格数据的新方法，通过CNN处理多模态数据，显著提高了分类性能，与传统方法相比具有本质区别。

关键设计：采用2D残差神经网络结构，设置了适当的损失函数，并通过提取8个感兴趣区域与其他特征（如病灶大小、初始症状严重性和恢复时间）进行联合训练。实验中对不同CNN架构进行了评估。

🖼️ 关键图片

img_0
img_1
img_2

📊 实验亮点

实验结果显示，基线逻辑回归模型在仅使用病灶大小时的分类准确率为0.678，加入初始症状严重性和恢复时间后分别提升至0.757和0.813。最终，结合8个感兴趣区域的2D残差神经网络达到了0.854的最高分类准确率，显示出显著的性能提升。

🎯 应用场景

该研究的潜在应用领域包括医疗影像分析和康复治疗，能够为中风患者的个性化康复方案提供数据支持。通过提高预测准确性，可能改善患者的康复效果和生活质量，具有重要的实际价值和社会影响。

📄 摘要（原文）

Machine learning offers great potential for automated prediction of post-stroke symptoms and their response to rehabilitation. Major challenges for this endeavour include the very high dimensionality of neuroimaging data, the relatively small size of the datasets available for learning, and how to effectively combine neuroimaging and tabular data (e.g. demographic information and clinical characteristics). This paper evaluates several solutions based on two strategies. The first is to use 2D images that summarise MRI scans. The second is to select key features that improve classification accuracy. Additionally, we introduce the novel approach of training a convolutional neural network (CNN) on images that combine regions-of-interest extracted from MRIs, with symbolic representations of tabular data. We evaluate a series of CNN architectures (both 2D and a 3D) that are trained on different representations of MRI and tabular data, to predict whether a composite measure of post-stroke spoken picture description ability is in the aphasic or non-aphasic range. MRI and tabular data were acquired from 758 English speaking stroke survivors who participated in the PLORAS study. The classification accuracy for a baseline logistic regression was 0.678 for lesion size alone, rising to 0.757 and 0.813 when initial symptom severity and recovery time were successively added. The highest classification accuracy 0.854 was observed when 8 regions-of-interest was extracted from each MRI scan and combined with lesion size, initial severity and recovery time in a 2D Residual Neural Network.Our findings demonstrate how imaging and tabular data can be combined for high post-stroke classification accuracy, even when the dataset is small in machine learning terms. We conclude by proposing how the current models could be improved to achieve even higher levels of accuracy using images from hospital scanners.