Physical Sciences and Mathematics Commons^™

Variable Importance And Prediction Methods For Longitudinal Problems With Missing Variables, Ivan Diaz, Alan E. Hubbard, Anna Decker, Mitchell Cohen

U.C. Berkeley Division of Biostatistics Working Paper Series

In this paper we present prediction and variable importance (VIM) methods for longitudinal data sets containing both continuous and binary exposures subject to missingness. We demonstrate the use of these methods for prognosis of medical outcomes of severe trauma patients, a field in which current medical practice involves rules of thumb and scoring methods that only use a few variables and ignore the dynamic and high-dimensional nature of trauma recovery. Well-principled prediction and VIM methods can thus provide a tool to make care decisions informed by the high-dimensional patient’s physiological and clinical history. Our VIM parameters can be causally interpreted …

Asymptotically Unbiased Estimator Of The Informational Energy With Knn, Angel Caţaron, Răzvan Andonie, Chinmei Y. Chueh

All Faculty Scholarship for the College of the Sciences

Motivated by machine learning applications (e.g., classification, function approximation, feature extraction), in previous work, we have introduced a non- parametric estimator of Onicescu’s informational energy. Our method was based on the k-th nearest neighbor distances between the n sample points, where k is a fixed positive integer. In the present contribution, we discuss mathematical properties of this estimator. We show that our estimator is asymptotically unbiased and consistent. We provide further experimental results which illustrate the convergence of the estimator for standard distributions.

Online Multi-Stage Deep Architectures For Feature Extraction And Object Recognition, Derek Christopher Rose

Doctoral Dissertations

Multi-stage visual architectures have recently found success in achieving high classification accuracies over image datasets with large variations in pose, lighting, and scale. Inspired by techniques currently at the forefront of deep learning, such architectures are typically composed of one or more layers of preprocessing, feature encoding, and pooling to extract features from raw images. Training these components traditionally relies on large sets of patches that are extracted from a potentially large image dataset. In this context, high-dimensional feature space representations are often helpful for obtaining the best classification performances and providing a higher degree of invariance to object transformations. …

Mobile Computing: Challenges And Opportunities For Autonomy And Feedback, Ole J. Mengshoel, Bob Iannucci, Abe Ishihara

Ole J Mengshoel

Using Methods From The Data-Mining And Machine-Learning Literature For Disease Classification And Prediction: A Case Study Examining Classification Of Heart Failure Subtypes, Peter C. Austin

Peter Austin

OBJECTIVE: Physicians classify patients into those with or without a specific disease. Furthermore, there is often interest in classifying patients according to disease etiology or subtype. Classification trees are frequently used to classify patients according to the presence or absence of a disease. However, classification trees can suffer from limited accuracy. In the data-mining and machine-learning literature, alternate classification schemes have been developed. These include bootstrap aggregation (bagging), boosting, random forests, and support vector machines.

STUDY DESIGN AND SETTING: We compared the performance of these classification methods with that of conventional classification trees to classify patients with heart failure (HF) …

Improved Cardiovascular Risk Prediction Using Nonparametric Regression And Electronic Health Record Data, Edward Kennedy, Wyndy Wiitala, Rodney Hayward, Jeremy Sussman

Edward H. Kennedy

Use of the electronic health record (EHR) is expected to increase rapidly in the near future, yet little research exists on whether analyzing internal EHR data using flexible, adaptive statistical methods could improve clinical risk prediction. Extensive implementation of EHR in the Veterans Health Administration provides an opportunity for exploration. Our objective was to compare the performance of various approaches for predicting risk of cerebrovascular and cardiovascular (CCV) death, using traditional risk predictors versus more comprehensive EHR data. Regression methods outperformed the Framingham risk score, even with the same predictors (AUC increased from 71% to 73% and calibration also improved). …