Community Spotlight: Sook-Lei Liew
My name is Sook-Lei Liew, but you can call me Lei. I am a Professor at the Chan Division of Occupational Science and Occupational Therapy at the University of Southern California. Because of the interdisciplinary nature of my work, I have the privilege of having joint appointments across Biokinesiology and Physical Therapy, Neurology, and Biomedical Engineering, and I am a member of the USC Stevens Neuroimaging and Informatics Institute. I serve as the Principal Investigator for DAPR. I love getting people together to collaborate, so I also lead the ENIGMA Stroke Recovery working group (https://enigma.ini.usc.edu/ongoing/enigma-stroke-recovery/), an international consortium of over 100 researchers working together to identify brain-based biomarkers of stroke recovery, and ReproRehab (https://www.reprorehab.usc.edu/), an educational training program to give rehabilitation researchers hands-on training in computer programming and open science tools. I’m only able to do these fun projects thanks to my amazing lab members (see the Neural Plasticity and Neurorehabilitation Lab: https://chan.usc.edu/npnl).
Tell us about your background
I was born in Missouri and grew up in Texas, and I double majored in English and Kinesiology at Rice University in Houston, Texas for undergrad. I didn’t know what I wanted to do when I graduated, so I took a career test, and it suggested I become an occupational therapist or a farmer. I chose OT! (But still hold dreams of living and working on a farm some day 😊). I came to the University of Southern California (USC) for my masters in OT. While there, I had my first neuroscience class, taught by Dr. Nancy Bagatell, and I was captivated by how much the brain controls our behaviors and even our personalities. I asked Dr. Bagatell if I should drop out of OT school and pursue neuroscience instead, but she (wisely) recommended I volunteer in a lab doing neuroscience research. I volunteered with Dr. Lisa Aziz-Zadeh, a new faculty member in our division, who used fMRI to study cognitive neuroscience, and I really loved it. I stayed on to complete my PhD in Occupational Science, with a concentration in Neuroscience. At that point, I had great clinical OT training and neuroscience research training, but I wanted to merge them. I then went to a postdoctoral fellowship at the National Institutes of Health with Dr. Leo Cohen to study the use of brain computer interfaces (BCI) and non-invasive brainstimulation (NIBS) for stroke recovery. While there, I also got to study under Dr. Surjo Soekadar (University of Tuebingen) and Dr. Pablo Celnik (Johns Hopkins University), which were all incredible experiences. Then I moved back to USC to start my faculty position, and the rest is history!
What sparked your interest in data science, machine learning, and artificial intelligence?
I started learning how to program Matlab around 2008, with the help of senior grad students who patiently answered my troubleshooting questions (thanks Tong and Dave!), as we needed specialized software to analyze fMRI data. I loved the challenge of programming; I felt like errors were puzzles and it was very rewarding to solve them! As I progressed in my career, I got involved with brainhacks (https://brainhack.org/), where I learned more of the uses for programming to make my data reproducible and open source, which led to many of the initiatives I work on now, like ReproRehab, which is an offshoot of ReproNim (https://repronim.org/).
What are you most excited about DAPR?
At the moment, I am most excited for DAPR to finish its first stage – I can’t wait until we have accomplished our initial mission of making it easy for people to combine and share their datasets! The work we have to do right now is largely “unsexy”; figuring out ways to easily bring researchers’ datasets together is tedious and time-consuming. But it is so, so critical because without larger, diverse datasets, we can’t apply AI to rehabilitation, and AI has so much potential to help us as clinicians do better for our patients (for more on this, see our latest paper on Collaborative AI for Precision Rehab). So I see our current work as a necessity that will enable way more fun and exciting things, like personalized rehabilitation.
What advice would you give to those who wish to follow a similar path?
Two things: (1) It sounds so hokey, but I really do just want people to find what makes them happy (that they can make a career out of). I’m not a great advance planner, and I could never have told you 10 or 20 years ago that this is where I would end up, but I have tried to follow what I’m interested in at each decision point in my career, and it’s been really rewarding. (2) Be willing to do the tedious hard stuff sometimes. These days especially, I think everyone’s looking for a new innovative AI tool or algorithm to do their tasks for them. But there is immense value also in slowly doing, and thoroughly understanding, the underlying work that you’re asking AI to do for you, so you can understand the intricacies and edge cases that AI may not catch.
Anything you are passionate about outside of DAPR?
I love my family! I love spending time with my husband and daughter and our extended family, and I wish someone would invent teleportation to help make seeing family easier. I also am a former ultra-marathoner and trail runner, now downgraded to an “aspirational” hobby runner (jogger/walker). 😊 I also like tennis, science fiction/fantasy books and shows, and awkward dramedy shows.
