Latent Gaussian process with composite likelihoods and numerical quadrature
This work addresses the challenge of dimensionality reduction for heterogeneous clinical data, which is incremental as it builds upon existing GPLVM methods.
The authors tackled the problem of learning low-dimensional representations from high-dimensional, heterogeneous clinical data with missing values by proposing an unsupervised generative model that extends the Gaussian process latent variable model (GPLVM) with multiple likelihoods and deep neural network back-constraints, achieving improved performance on a standard benchmark and Parkinson's disease clinical data.
Clinical patient records are an example of high-dimensional data that is typically collected from disparate sources and comprises of multiple likelihoods with noisy as well as missing values. In this work, we propose an unsupervised generative model that can learn a low-dimensional representation among the observations in a latent space, while making use of all available data in a heterogeneous data setting with missing values. We improve upon the existing Gaussian process latent variable model (GPLVM) by incorporating multiple likelihoods and deep neural network parameterised back-constraints to create a non-linear dimensionality reduction technique for heterogeneous data. In addition, we develop a variational inference method for our model that uses numerical quadrature. We establish the effectiveness of our model and compare against existing GPLVM methods on a standard benchmark dataset as well as on clinical data of Parkinson's disease patients treated at the HUS Helsinki University Hospital.