CS 7626: Behavioral Imaging
Introduction to Behavioral Imaging
CS 7626 and CS 4803-IBI combined sections
The most up-to-date information on Covid-19 is on the TECH Moving Forward (Links to an external site.) website and in the Academic Restart Frequently Asked Questions (Links to an external site.). If you have not tested positive but are ill or have been exposed to someone who is ill, please follow the Covid-19 Exposure Decision Tree (Links to an external site.) for reporting your illness.
If you are on campus this semester, please consider being tested regularly. Information on free, voluntary testing on campus can be found here: https://mytest.gatech.edu/
Some disruption to classes or services is inevitable, but Georgia Tech is making every effort to ensure continuity of operations. As is the case in any semester, faculty may cancel a class if they have an illness or emergency situation and cover any missed material at their own discretion. If an instructor needs to cancel a class, they will notify students as early as possible.
If you are ill and unable to do course work this will be treated similarly to any student illness. The Dean of Students will have been contacted when you report your positive test and will notify your instructor that you may be unable to attend class events or finish your work as the result of a health issue. Your instructor will not be told the reason. We have asked all faculty to be lenient and understanding when setting work deadlines or expecting students to finish work, and so you should be able to catch up with any work that you miss while in quarantine or isolation. Your instructor may make available any video recordings of classes or slides that have been used while you are absent, and may prepare some complementary asynchronous assignments that compensate for your inability to participate in class sessions. Ask your instructor for the details.
These uncertain times can be difficult, and many students may need help in dealing with stress and mental health. The CARE Center (Links to an external site.) and the Counseling Center (Links to an external site.), and Stamps Health Services (Links to an external site.) will offer both in-person and virtual appointments. Face-to-face appointments will require wearing a face covering and social distancing, with exceptions for medical examinations. Student Center services and operations are available on the Student Center (Links to an external site.) website. For more information on these and other student services, contact the Vice President and Dean of Students or the Division of Student Life (Links to an external site.).
James M. Rehg (you can call me Jim)
Office Hours: Thursdays 9:00-10:00am and by appointment
Office Hours Location: MS Teams (link coming soon)
Course Time and Location
This course will be 100% remote and lectures will be synchronous. All lectures will be made available in a recorded format. There is no attendance requirement.
Classes will be held on Mondays, Wednesdays, and Fridays from 3:30pm-4:30pm Eastern Time Zone. Lectures will meet on Bluejeans at the following link: https://bluejeans.com/203011482
All slides will be linked into the lecture schedule at the end of this syllabus.
Asking Questions During Lectures
Since our course size is modest we will use Bluejeans meetings for our initial class meetings. This will allow us to interact directly. I am investigating moving to Microsoft Teams for our lectures and will post more on this as it develops.
The TA for this course is Shivam Khare (firstname.lastname@example.org)
Office hours will be held (Wedneday 2:30-3:30 pm)
Check the calendar for the schedule of TA Office Hours (https://bluejeans.com/592632758)
This course will provide an introduction to Behavioral Imaging, a new research field which encompasses the measurement, modeling, analysis, and visualization of behaviors from multi-modal sensor data. It is tailored for undergraduate and graduate students who are interested in this emerging field. The course is designed to provide:
- A broad introduction to research questions in Behavioral Imaging (BI)
- An in-depth understanding of the key technologies for BI
- An overview of the psychology literature relating to behavior from a computational perspective
- Hands-on experience in working with relevant sensor data
Beginning in infancy, individuals acquire the social and communicative skills which are vital for a healthy and productive life, through face-to-face interactions with caregivers and peers. However, children with developmental delays face great challenges in acquiring these skills, resulting in substantial lifetime risks. A goal of research in Behavioral Imaging is to develop computational methods that can support the fine-grained and large-scale measurement and analysis of social and developmental behaviors, with the potential to positively impact the diagnosis and treatment of developmental disorders such as autism.
Clinical health domains can also benefit from a computational study of behavior. A key example is chronic health conditions, such as heart disease, asthma, and diabetes, in which health-related behaviors such as smoking or unhealthy eating habits play a critical role. The ability to treat chronic conditions is hampered by an inability to reliably measure health-related behaviors, particularly in naturalistic (field) conditions. Advances in wearable sensing technologies, known as mobile health or mHealth, have the potential to enable a new data-driven approach to the treatment of such conditions. This course attempts to bridge the gap between multiple disciplines, such as mHealth, computational psychology, developmental psychology, and behavioral medicine, which are linked by a common interest in computational methods for understanding and measuring the behavioral underpinnings of adverse health outcomes.
A key goal of this course is to equip data science researchers with the means to access the relevant literature in psychology, physiology, and psychophysiology which provide the theoretical underpinning for most health-related technology development. It is still too common that researchers gain access to health-related datasets and apply machine learning methods without a sufficiently clear understanding of how the data relates to health outcomes. Real-world improvements in health outcomes, which is the promise of advanced ML and sensor technologies, can only arise through a deep collaboration between technologists and domain experts, which this course aspires to enable.
This class will be self-contained with respect to the core concepts and class material. Students should have familiarity with machine learning at the level of the undergraduate ML class CS 4641. If you do not have basic familiarity with classical machine learning methods such as support vector machines or random forest classifiers, then it will be challenging to read the papers and do a final project. The technical papers will also require some familiarity with deep learning in order to understand the readings, but you do not have to have an extensive deep learning programming background to take this class, since it is still possible to use classical ML methods in health projects (particularly since existing datasets still tend to be small). Background material in psychology, physiology, and related topics will be provided, so there are no subject-matter prerequisites. The projects will require facility with Python and the basics of machine learning such as the use of standard libraries like scikit-learn. Tutorial material will be provided for more advanced content.
As this is a new research area, there are no textbooks that cover the breadth of this course. Recommended readings will be provided for all lectures and class discussions. The following text is a useful reference for the mobile health content of the course, and some chapters will be provided in the readings:
- Mobile Health: Sensors, Analytic Methods, and Applications, Editors: Rehg, JM, Murphy, S, and Kumar, S. Springer 2017
- Canvas (here): For syllabus, projects and quizzes, and critical announcements.
- GradeScope (LINK TBD): For project submission and grading
- Piazza: For general questions and discussions.
- Teams (LINK TBD): For TA and instructor office hours.
- Bluejeans: For class lectures
The course material is divided into units, and each unit will feature some combination of lectures on technology and topics in psychology and physiology, along with discussion periods covering research papers from the technology and psychology literatures. Each discussion period will be led by a team of students and will cover 2 papers, one from the technology literature and one from the behavioral literature. This task supports the goal of gaining experience in reading the original literature in psychology and health. All students are expected to read the papers before each meeting and participate in the class discussion. Each student will make two presentations during the semester as part of their team. A quiz will be given at the end of each unit, which will test the core concepts in psychology, physiology, and related domains that have been covered. There will be a final project for which the class will work in teams and will address one area of BI technology.
- 2 In-Class Presentations: 70% (35% x 2)
- 3 Quizzes: 30%
The quizzes will be take home, and you can work on them at any time and take as long as want to complete them, as long as you submit your answers by the deadlines listed. However you may only submit your answers once. We will not have class the day the quiz is released so that you can use the class period to take the quiz if you'd prefer. The quiz must be your own work, you are not allowed to collaborate on the quiz or discuss it with anyone.
Attendance will not be recorded and will not be used in assessing grades.
Final Project: This was eliminated from the syllabus due to the difficulty of making individualized projects work under the current remote/covid conditions.
I reserve the right to modify any of these plans as need be during the course of the class; however, I won't do anything too drastic, and you'll be informed as far in advance as possible.
You must abide by the academic honor code of Georgia Tech.