Radu Stefan Niculescu

Learning to Decode Cognitive States from Brain Images

Over the past decade, functional Magnetic Resonance Imaging (fMRI) has emerged as a powerful new instrument to collect vast quantities of data about activity in the human brain. A typical fMRI experiment can produce a three-dimensional image related to the human subjects brain activity every half second, at a spatial resolution of a few millimeters. As in other modern empirical sciences, this new instrumentation has led to a flood of new data, and a corresponding need for new data analysis methods. We describe recent research applying machine learning methods to the problem of classifying the cognitive state of a human subject based on fRMI data observed over a single time interval. In particular, we present case studies in which we have successfully trained classifiers to distinguish cognitive states such as (1) whether the human subject is looking at a picture or a sentence, (2) whether the subject is reading an ambiguous or non-ambiguous sentence, and (3) whether the word the subject is viewing is a word describing food, people, buildings, etc. This learning problem provides an interesting case study of classifier learning from extremely high dimensional (105 features), extremely sparse (tens of training examples), noisy data. This paper summarizes the results obtained in these three case studies, as well as lessons learned about how to successfully apply machine learning methods to train classifiers in such settings.

Data mining for improved cardiac care

Cardiovascular Disease (CVD) is the single largest killer in the world. Although, several CVD treatment guidelines have been developed to improve quality of care and reduce healthcare costs, for a number of reasons, adherence to these guidelines remains poor. Further, due to the extremely poor quality of data in medical patient records, most of todays healthcare IT systems cannot provide significant support to improve the quality of CVD care (particularly in chronic CVD situations which contribute to the majority of costs).We present REMIND, a Probabilistic framework for Reliable Extraction and Meaningful Inference from Nonstructured Data. REMIND integrates the structured and unstructured clinical data in patient records to automatically create high-quality structured clinical data. There are two principal factors that enable REMIND to overcome the barriers associated with inference from medical records. First, patient data is highly redundant -- exploiting this redundancy allows us to deal with the inherent errors in the data. Second, REMIND performs inference based on external medical domain knowledge to combine data from multiple sources and to enforce consistency between different medical conclusions drawn from the data -- via a probabilistic reasoning framework that overcomes the incomplete, inconsistent, and incorrect nature of data in medical patient records.This high-quality structuring allows existing patient records to be mined to support guideline compliance and to improve patient care. However, once REMIND is configured for an institutions data repository, many other important clinical applications are also enabled, including: quality assurance; therapy selection for individual patients; automated patient identification for clinical trials; data extraction for research studies; and to relate financial and clinical factors. REMIND provides value across the continuum of healthcare, ranging from small physician practice databases to the most complex hospital IT systems, from acute cardiac care to chronic CVD management, and to experimental research studies. REMIND is currently deployed across multiple disease areas over a total of 5,000,000 patients across the US.

Modeling fMRI data generated by overlapping cognitive processes with unknown onsets using Hidden Process Models

We present a new method for modeling fMRI time series data called Hidden Process Models (HPMs). Like several earlier models for fMRI analysis, Hidden Process Models assume that the observed data is generated by a sequence of underlying mental processes that may be triggered by stimuli. HPMs go beyond these earlier models by allowing for processes whose timing may be unknown, and that might not be directly tied to specific stimuli. HPMs provide a principled, probabilistic framework for simultaneously learning the contribution of each process to the observed data, as well as the timing and identities of each instantiated process. They also provide a framework for evaluating and selecting among competing models that assume different numbers and types of underlying mental processes. We describe the HPM framework and its learning and inference algorithms, and present experimental results demonstrating its use on simulated and real fMRI data. Our experiments compare several models of the data using cross-validated data log-likelihood in an fMRI study involving overlapping mental processes whose timings are not fully known.

Automatic medical coding of patient records via weighted ridge regression

In this paper, we apply weighted ridge regression to tackle the highly unbalanced data issue in automatic large-scale ICD-9 coding of medical patient records. Since most of the ICD-9 codes are unevenly represented in the medical records, a weighted scheme is employed to balance positive and negative examples. The weights turn out to be associated with the instance priors from a probabilistic interpretation, and an efficient EM algorithm is developed to automatically update both the weights and the regularization parameter. Experiments on a large-scale real patient database suggest that the weighted ridge regression outperforms the conventional ridge regression and linear support vector machines (SVM).