Etiologies and Impact of Readmission Rates in the First 180 Days After Hematopoietic Stem Cell Transplantation in Children, Adolescents, and Young Adults.

INTRODUCTION: High rates of patients require readmission to the hospital within 6 months of hematopoietic stem cell transplantation (HSCT). We investigated the relationship between readmission rates and outcomes after HSCT in children, adolescents, and young adults (CAYA). MATERIALS AND METHODS: A retrospective analysis of patients (26 years or younger) treated with HSCT was conducted. RESULTS: A chart review of 435 CAYA who underwent HSCT from 2008 to 2015 revealed that 171 patients (39%) had at least 1 hospital readmission within 180 days of transplant; 87% received allogeneic and 13% received autologous HSCT. A total of 312 readmission events were reported. The median follow-up time was 31 months. Documented infection (n=99) and graft-versus-host disease complications (n=60) were the most common causes. Higher than 2 readmission rates were associated with lower overall survival (OS) (P=0.001) and disease-free survival (P<0.001) in patients who received allogeneic HSCT. These findings were not found in the autologous HSCT. In a multivariate analysis of those who received allogeneic HSCT, prior treatment with ≥2 chemotherapy regimens (P=0.03) was independent predictor of lower OS. There were also trends noted toward lower OS for patients with documented infections at index admission or subsequent readmissions (P=0.09). CONCLUSIONS: More than 2 hospital readmissions within 6 months of allogeneic HSCT in CAYA, who are either heavily pretreated or had documented infections at index admission or subsequent readmissions adversely affected the outcomes.

Manifold learning for object tracking with multiple nonlinear models.

This paper presents a novel manifold learning algorithm for high-dimensional data sets. The scope of the application focuses on the problem of motion tracking in video sequences. The framework presented is twofold. First, it is assumed that the samples are time ordered, providing valuable information that is not presented in the current methodologies. Second, the manifold topology comprises multiple charts, which contrasts to the most current methods that assume one single chart, being overly restrictive. The proposed algorithm, Gaussian process multiple local models (GP-MLM), can deal with arbitrary manifold topology by decomposing the manifold into multiple local models that are probabilistic combined using Gaussian process regression. In addition, the paper presents a multiple filter architecture where standard filtering techniques are integrated within the GP-MLM. The proposed approach exhibits comparable performance of state-of-the-art trackers, namely multiple model data association and deep belief networks, and compares favorably with Gaussian process latent variable models. Extensive experiments are presented using real video data, including a publicly available database of lip sequences and left ventricle ultrasound images, in which the GP-MLM achieves state of the art results.