My research is at the intersection of health, mobile and ubiquitous computing, AI, and Human Computer Interaction.
I work on computational modeling of human behavior from data streams collected through mobile, wearable, and embedded sensors. Examples of my work in the health domain include detection of behavior change in people with depression, predicting mania-depression episodes in bipolar disorder, estimation of symptom severity in cancer patients, and modeling of surgical activities inside the operating room.
I also work on intelligent applications for social good. My recent work in Context-aware Peer-to-Peer economic exchange is focused on connecting communities of people through mobile technology to enable successful and meaningful service transactions, especially in low-income communities. In my research, I draw on methods from Machine Learning, Data Mining, Statistics, and Human-Computer Interaction.
Seeking PhD Researchers
The Intelligent Human-Centered Computing (IHCC) lab at the University of Virginia (https://engineering.virginia.edu/intelligent-human-centered-computing) is looking for PhD students and postdocs with a strong background in Computer Science, Math, Machine Learning, and AI to work on research projects in the area of ubiquitous and mobile computing and human-centered computing focused on social good and health.
The IHCC lab aims to intelligently explore the patterns inherent to human behavior through machine learning and computational models of passive data streams. From this knowledge, we can make predictions about potential health outcomes and suggest individualized improvements to promote a healthier, happier community.
TO APPLY: Please contact Dr. Afsaneh Doryab (email@example.com) with qualifications and to discuss research fit.
News and Events
AAAI 2020 Conference
February 7th-12th, 2020
I presented my research in collaboration with Xi Chen at the 34th AAAI Conference on Artificial Intelligence in New York, New York. The paper proposes a novel approach to improving the accuracy of feature selection, specifically in data sets with large feature sets and a small number of data points. Existing methods of feature selection typically do not achieve high accuracy. The approach presented optimizes feature selection through Frequent Pattern Growth algorithm to identify frequently occurring sets that appear among the top features selected.
Doryab Presents the First Paper on Modeling Biobehavioral Rhythms for Predicting Rehospitalization Risk
ESE Weekly News | September 27th, 2019
The paper titled "Modeling Biobehavioral Rhythms with Passive Sensing in the Wild: A Case Study to Predict Readmission Risk after Pancreatic Surgery" was presented at Ubicomp, the Association for Computing Machinery (ACM) leading conference on ubiquitous and mobile computing.
Doryab's Paper on the Estimation of Symptom Severity During Chemotherapy from Passively Sensed Data Selected as the Best Cancer Informatics Paper by International Medical Informatics Association
ESE Weekly News | September 20th, 2019
The Editorial Board of the 2019 IMIA Yearbook of Medical Informatics selected article entitled: "Estimation of symptom severity during chemotherapy from passively sensed data: Exploratory study" which was led by Dr. Afsaneh Doryab for listing in the 2019 edition of the Yearbook as one of the best articles published in 2018 in the ‘Cancer Informatics’ subfield of medical informatics.
Chen and Doryab's poster presented at the 34th AAAI Conference on Artificial Intelligence