IEEE J Biomed Health Inform. 2023 Nov;27(11):5357-5368. doi: 10.1109/JBHI.2023.3312597. Epub 2023 Nov 7.
This work considers the problem of segmenting heart sounds into their fundamental components. We unify statistical and data-driven solutions by introducing Markov-based Neural Networks (MNNs), a hybrid end-to-end framework that exploits Markov models as statistical inductive biases for an Artificial Neural Network (ANN) discriminator. We show that an MNN leveraging a simple one-dimensional Convolutional ANN significantly outperforms two recent purely data-driven solutions for this task in two publicly available datasets: PhysioNet 2016 (Sensitivity: 0.947 ±0.02; Positive Predictive Value : 0.937 ±0.025) and the CirCor DigiScope 2022 (Sensitivity: 0.950 ±0.008; Positive Predictive Value: 0.943 ±0.012). We also propose a novel gradient-based unsupervised learning algorithm that effectively makes the MNN adaptive to unseen datum sampled from unknown distributions. We perform a cross dataset analysis and show that an MNN pre-trained in the CirCor DigiScope 2022 can benefit from an average improvement of 3.90% Positive Predictive Value on unseen observations from the PhysioNet 2016 dataset using this method.
这项工作考虑了将心音分解为其基本成分的问题。我们通过引入基于马尔可夫的神经网络（MNN），将统计和数据驱动的解决方案统一起来，这是一种端到端的混合框架，利用马尔可夫模型作为人工神经网络（ANN）鉴别器的统计归纳偏差。我们表明，利用简单的一维卷积 ANN 的 MNN 在两个公开可用的数据集上显著优于这一任务的两个最近的纯数据驱动解决方案：PhysioNet 2016（灵敏度：0.947 ±0.02；阳性预测值：0.937 ±0.025）和 CirCor DigiScope 2022（灵敏度：0.950 ±0.008；阳性预测值：0.943 ±0.012）。我们还提出了一种新颖的基于梯度的无监督学习算法，该算法有效地使 MNN 能够适应来自未知分布的未见过的数据。我们进行了跨数据集的分析，并表明，使用这种方法，在 CirCor DigiScope 2022 中预先训练的 MNN 可以从 PhysioNet 2016 数据集中未见过的观测值中受益，平均阳性预测值提高 3.90%。
