Graphical methods for understanding changes in states: Understanding medication use pathways.

OBJECTIVES: As epidemiological studies become longer and larger, the field needs novel graphical methods to visualize complex longitudinal data. The aim of this study was to present the Slinkyplot, a longitudinal crosstabulation, to illustrate patterns of antidepressant use in a large prospective cohort of older adults with mild cognitive impairment. METHODS: Data from the National Alzheimer's Coordinating Center are used to track switches between different states and types of antidepressant use. A Slinkyplot is populated with rows representing the state of medication use at each timepoint and columns representing the state at each subsequent visit. RESULTS: The constructed Slinkyplots display the common practice of switching on and off different antidepressants over time, with citalopram, sertraline, and bupropion most commonly used followed by switching to another SSRI or SNRI as second-line treatment. CONCLUSIONS: Slinkyplots are an innovative graphical means of visualizing complex patterns of transitions between different states over time for large longitudinal studies.

rConverse: Moment by Moment Conversation Detection Using a Mobile Respiration Sensor.

Monitoring of in-person conversations has largely been done using acoustic sensors. In this paper, we propose a new method to detect moment-by-moment conversation episodes by analyzing breathing patterns captured by a mobile respiration sensor. Since breathing is affected by physical and cognitive activities, we develop a comprehensive method for cleaning, screening, and analyzing noisy respiration data captured in the field environment at individual breath cycle level. Using training data collected from a speech dynamics lab study with 12 participants, we show that our algorithm can identify each respiration cycle with 96.34% accuracy even in presence of walking. We present a Conditional Random Field, Context-Free Grammar (CRF-CFG) based conversation model, called rConverse, to classify respiration cycles into speech or non-speech, and subsequently infer conversation episodes. Our model achieves 82.7% accuracy for speech/non-speech classification and it identifies conversation episodes with 95.9% accuracy on lab data using a leave-one-subject-out cross-validation. Finally, the system is validated against audio ground-truth in a field study with 32 participants. rConverse identifies conversation episodes with 71.7% accuracy on 254 hours of field data. For comparison, the accuracy from a high-quality audio-recorder on the same data is 71.9%.

Learning Time Series Detection Models from Temporally Imprecise Labels.

In this paper, we consider the problem of learning time series detection models from temporally imprecise labels. In this problem, the data consist of a set of input time series, and supervision is provided by a sequence of noisy time stamps corresponding to the occurrence of positive class events. Such temporally imprecise labels occur in areas like mobile health research when human annotators are tasked with labeling the occurrence of very short duration events. We propose a general learning framework for this problem that can accommodate different base classifiers and noise models. We present results on real mobile health data showing that the proposed framework significantly outperforms a number of alternatives including assuming that the label time stamps are noise-free, transforming the problem into the multiple instance learning framework, and learning on labels that were manually aligned.

Hierarchical Span-Based Conditional Random Fields for Labeling and Segmenting Events in Wearable Sensor Data Streams.

The field of mobile health (mHealth) has the potential to yield new insights into health and behavior through the analysis of continuously recorded data from wearable health and activity sensors. In this paper, we present a hierarchical span-based conditional random field model for the key problem of jointly detecting discrete events in such sensor data streams and segmenting these events into high-level activity sessions. Our model includes higher-order cardinality factors and inter-event duration factors to capture domain-specific structure in the label space. We show that our model supports exact MAP inference in quadratic time via dynamic programming, which we leverage to perform learning in the structured support vector machine framework. We apply the model to the problems of smoking and eating detection using four real data sets. Our results show statistically significant improvements in segmentation performance relative to a hierarchical pairwise CRF.

Towards collaborative filtering recommender systems for tailored health communications.

The goal of computer tailored health communications (CTHC) is to promote healthy behaviors by sending messages tailored to individual patients. Current CTHC systems collect baseline patient "profiles" and then use expert-written, rule-based systems to target messages to subsets of patients. Our main interest in this work is the study of collaborative filtering-based CTHC systems that can learn to tailor future message selections to individual patients based explicit feedback about past message selections. This paper reports the results of a study designed to collect explicit feedback (ratings) regarding four aspects of messages from 100 subjects in the smoking cessation support domain. Our results show that most users have positive opinions of most messages and that the ratings for all four aspects of the messages are highly correlated with each other. Finally, we conduct a range of rating prediction experiments comparing several different model variations. Our results show that predicting future ratings based on each user's past ratings contributes the most to predictive accuracy.