Notice: This lecture is a research seminar for credit.
September 26th, Wednesday, 2018 2:30-4:00 p.m.
People often talk about
trade-off between model accuracy and interpretability. However, in healthcare,
we need both. In this talk, I will present two examples of sophisticated models
that can be accurate yet interpretable.
Intensive Care Unit (ICU) outcome prediction: Integration
of high-density ICU monitoring data with the discrete clinical events
(including diagnosis, medications, labs) is challenging but potentially
rewarding since richness and granularity in such multimodal data increase the
possibilities for accurate detection of complex problems and predicting
outcomes (e.g., length of stay and mortality). We propose Recurrent Attentive
and Intensive Model (RAIM) for jointly analyzing continuous monitoring data and
discrete clinical events. RAIM introduces an efficient attention mechanism for
continuous monitoring data (e.g., ECG), which is guided by discrete clinical
events (e.g, medication usage).
Heart Failure Phenotyping: Understanding subtypes
of heart failure patients is extremely hard but important. We propose an
integer tensor factorization method SUSTain to model electronic health
records. EHR data are commonly represented by integers (e.g., the values
correspond to event counts or ordinal measures). The conventional approach is
to treat integer data as real, and then apply real-valued factorizations.
However, doing so fails to preserve important characteristics of the original
data, thereby making it hard to interpret the results. In our preliminary
study, 87% of the resulting phenotypes were clinically meaningful.
1.RAIM: Recurrent Attentive and Intensive Modeling of Multimodal
Continuous Patient Monitoring Data, KDD’18
2.SUSTain: Scalable Unsupervised Scoring for Tensors and its
Application to Phenotyping, KDD’18
Sun is an Associate Professor of College of Computing at Georgia Tech. Prior to
Georgia Tech, he was a researcher at IBM TJ Watson Research Center. His
research focuses on health analytics and data mining, especially in designing
tensor factorizations, deep learning methods, and large-scale predictive
modeling systems. Dr. Sun has been collaborating with many healthcare
published over 120 papers and filed over 20 patents (5 granted). He has
received SDM/IBM early
career research award 2017, ICDM best research paper award in
2008, SDM best research paper award in 2007, and KDD Dissertation runner-up
award in 2008. Dr. Sun received B.S. and M.Phil. in Computer Science from Hong
Kong University of Science and Technology in 2002 and 2003, PhD in Computer
Science from Carnegie Mellon University in 2007 advised by Christos Faloutsos.
Professors and students of TBSI are welcome to attend. The lecture is also open to the public. For off-campus personnel, please scan the QR code and and fill in your information (name, company, contact number, ID number). The language of the lecture is English.