GeoAI Symposium: Advanced Analytical Methods and Sensor Data Analytics for Health Research

Type: Paper
Theme:
Sponsor Groups: Health and Medical Geography Specialty Group, Cyberinfrastructure Specialty Group, Spatial Analysis and Modeling Specialty Group
Organizers: Alexander Hohl, Michael Desjardins, Eun-Kyeong Kim
Chairs: Eun-Kyeong Kim

Call for Submissions

If you are interested in joining this session, please email your PIN, title, abstract, and contact information to the following organizers:

Eun-Kyeong Kim (University of Zurich, eun-kyeong.kim@geo.uzh.ch);
Alexander Hohl (University of Utah; alexander.hohl@geog.utah.edu);
Michael R. Desjardins (UNC-Charlotte; mdesjar2@uncc.edu);
Claudio Owusu (UNC-Charlotte; cowusu@uncc.edu).

Please note that we will not accept submissions after the AAG deadline.


Description

This session is part of the 3rd Symposium on GeoAI and Deep Learning for Geospatial Research as well as Symposium on Advances in Computational Approaches for Geospatial Health Applications.

Description:
Advanced computational capabilities that emerged through continuous technological and methodological developments have transformed many scientific disciplines, including the domain of health and medical geography. Today, scientists tackle computational challenges that used to be virtually impossible to solve, because 1) our ability to collect and store health-related data has improved substantially, and 2) analytical methods for solving scientific problems can now be applied on a massive scale. Therefore, the spatiotemporal analysis and modelling of health-related issues has experienced and driven fundamental changes.

In the last decade, advanced analytical methods and data-driven analytics have made progress in health domain. Especially advances in applied machine learning (ML), sensor data analytics, and computational methods for large and heterogeneous datasets are remarkable, thanks to an abundance of available data, improved computational power, and sensing techniques. ML enables automatic detection and/or highly accurate prediction of various infectious/environmental diseases and mental/physical conditions. Sensor technologies enable capturing the physical and mental health status in real life and analyzing them at a finer temporal and geographical resolution than ever before. Massive data is produced from wearable sensors as well as ambulatory/ momentary assessments. This trend opens new research opportunities to delve into diverse aspects of individuals’ health from geographical perspectives in an objective manner.

This session seeks for participants who are applying advanced ML, sensing techniques, and computational methods to geospatial health research. The potential topics include, but are not limited to:
1) Applied ML in the health domain;
2) Automated disease detection and diagnostics;
3) Sensor data-driven geospatial health research;
4) Health research based on ambulatory/momentary assessments;
5) Physical activities and physical/mental health;
6) Wearable/mobile sensor data processing, feature engineering, and visualization;
7) High-performance computing and spatiotemporal data mining to analyze complex and massive spatiotemporal data for health research;
8) Interpretability of ML in health research.

Discussants: Ying Song (University of Minnesota) and Wei Luo (Harvard Medical School).
Sponsor Groups: Geographic Information Science and Systems Specialty Group, Cyberinfrastructure Specialty Group, Spatial Analysis and Modeling Specialty Group, Health and Medical Geography Specialty Group
To present your research in our session, please submit your abstract through the AAG website (http://www.aag.org/cs/annualmeeting/register) and send your PIN to the session organizers listed above.


Agenda

Type Details Minutes
Presenter Changzhen Wang*, Louisiana State University, Fahui Wang, Louisiana State University, Tracy Onega, Geisel School of Medicine at Dartmouth, Network Optimization Approach to Delineating Health Care Service Areas: From Spatially Constrained Louvain to Leiden Algorithm 15

To access contact information login