Dr. Emily Huang

Assistant Professor

Dr. Huang photo, decorative

 

 

 

 

 

Dr. Huang’s

  • BSE University of Pennsylvania '12 -- Bioengineering and Mathematics
  • PhD Johns Hopkins University '17 -- Biostatistics
  • Digital phenotyping, Causal inference, Clinical trials
  • My research addresses novel questions and develops new approaches in causal inference and digital phenotyping. My goal in causal inference is to go beyond the average treatment effect and investigate the question of the fraction who benefit from treatment in randomized trial analyses. Since the fraction who benefit is generally a non-identifiable parameter, my collaborators and I have derived its identifiable lower and upper bounds without parametric assumptions. We also extended the formulation to incorporate support restrictions that further tighten the bounds. From this formulation, we developed the first consistent estimator of the bounds for ordinal outcomes with support restrictions. More recently, we have developed a new confidence interval method for the fraction who benefit, which is advantageous to existing ones in its guaranteed consistency.

    In addition to causal inference, I also work in the emerging field of digital phenotyping, which is the moment-by-moment quantification of the individual-level human phenotype in situ using data from personal digital devices. We analyze patients’ smartphone data (such as data collected by the accelerometer and GPS sensors in the phone) to monitor patients’ health status. My current project aims to infer physical activity levels from the acceleration and angular velocity of smartphone movement in near real-time. I have designed and conducted my own IRB-approved field study in which participants performed various activities while wearing smartphones. Using the data set, I am developing new methods to perform activity recognition from accelerometer and gyroscope (angular velocity) measurements.

Professional Website

  • Constructing a Confidence Interval for the Fraction who Benefit from Treatment, Using Randomized Trial Data
  • Modeling of Clinical Phenotypes Assessed at Discrete Study Visits
  • Safety and Efficacy of Minimally Invasive Surgery plus Alteplase in Intracerebral Haemorrhage Evacuation (MISTIE): a Randomized, Controlled, Open-label, Phase 2 Trial
  • Inequality in Treatment Benefits: Can We Determine if a New Treatment Benefits the Many or the Few?
Introductory statistics, Design and Sampling
Piano, clarinet, reading

 

 

 

Comments are closed.