Ethnicity influences phenotype and clinical outcomes: Comparing a South American with a North American inflammatory bowel disease cohort.

Inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn disease (CD), has emerged as a global disease with an increasing incidence in developing and newly industrialized regions such as South America. This global rise offers the opportunity to explore the differences and similarities in disease presentation and outcomes across different genetic backgrounds and geographic locations. Our study includes 265 IBD patients. We performed an exploratory analysis of the databases of Chilean and North American IBD patients to compare the clinical phenotypes between the cohorts. We employed an unsupervised machine-learning approach using principal component analysis, uniform manifold approximation, and projection, among others, for each disease. Finally, we predicted the cohort (North American vs Chilean) using a random forest. Several unsupervised machine learning methods have separated the 2 main groups, supporting the differences between North American and Chilean patients with each disease. The variables that explained the loadings of the clinical metadata on the principal components were related to the therapies and disease extension/location at diagnosis. Our random forest models were trained for cohort classification based on clinical characteristics, obtaining high accuracy (0.86 = UC; 0.79 = CD). Similarly, variables related to therapy and disease extension/location had a high Gini index. Similarly, univariate analysis showed a later CD age at diagnosis in Chilean IBD patients (37 vs 24; P = .005). Our study suggests a clinical difference between North American and Chilean IBD patients: later CD age at diagnosis with a predominantly less aggressive phenotype (39% vs 54% B1) and more limited disease, despite fewer biological therapies being used in Chile for both diseases.

Predicting nationwide obesity from food sales using machine learning.

The obesity epidemic progresses everywhere across the globe, and implementing frequent nationwide surveys to measure the percentage of obese population is costly. Conversely, country-level food sales information can be accessed inexpensively through different suppliers on a regular basis. This study applies a methodology to predict obesity prevalence at the country-level based on national sales of a small subset of food and beverage categories. Three machine learning algorithms for nonlinear regression were implemented using purchase and obesity prevalence data from 79 countries: support vector machines, random forests and extreme gradient boosting. The proposed method was validated in terms of both the absolute prediction error and the proportion of countries for which the obesity prevalence was predicted satisfactorily. We found that the most-relevant food category to predict obesity is baked goods and flours, followed by cheese and carbonated drinks.

Compositionally-warped Gaussian processes.

The Gaussian process (GP) is a nonparametric prior distribution over functions indexed by time, space, or other high-dimensional index set. The GP is a flexible model yet its limitation is given by its very nature: it can only model Gaussian marginal distributions. To model non-Gaussian data, a GP can be warped by a nonlinear transformation (or warping) as performed by warped GPs (WGPs) and more computationally-demanding alternatives such as Bayesian WGPs and deep GPs. However, the WGP requires a numerical approximation of the inverse warping for prediction, which increases the computational complexity in practice. To sidestep this issue, we construct a novel class of warpings consisting of compositions of multiple elementary functions, for which the inverse is known explicitly. We then propose the compositionally-warped GP (CWGP), a non-Gaussian generative model whose expressiveness follows from its deep compositional architecture, and its computational efficiency is guaranteed by the analytical inverse warping. Experimental validation using synthetic and real-world datasets confirms that the proposed CWGP is robust to the choice of warpings and provides more accurate point predictions, better trained models and shorter computation times than WGP.

Multikernel least mean square algorithm.

The multikernel least-mean-square algorithm is introduced for adaptive estimation of vector-valued nonlinear and nonstationary signals. This is achieved by mapping the multivariate input data to a Hilbert space of time-varying vector-valued functions, whose inner products (kernels) are combined in an online fashion. The proposed algorithm is equipped with novel adaptive sparsification criteria ensuring a finite dictionary, and is computationally efficient and suitable for nonstationary environments. We also show the ability of the proposed vector-valued reproducing kernel Hilbert space to serve as a feature space for the class of multikernel least-squares algorithms. The benefits of adaptive multikernel (MK) estimation algorithms are illuminated in the nonlinear multivariate adaptive prediction setting. Simulations on nonlinear inertial body sensor signals and nonstationary real-world wind signals of low, medium, and high dynamic regimes support the approach.