Dr. Xueying Zhao is an R&D engineer at Becton, Dickinson and Company (BD), where she integrates device innovation with AI-driven health care solutions.

After earning a Ph.D. in chemical engineering from the University of Virginia, her passion for solving and streamlining health care problems led her to pursue a second master’s degree in computer science at the Georgia Institute of Technology. This multidisciplinary expertise positions her at the forefront of health care innovation, demonstrating the power of continuous learning and interdisciplinary thinking.

Dr. Zhao is committed to sharing AI’s profound potential for impact in the health care space and has given talks on AI in health care at WE24 in Chicago and the 2024 WE Local conference in Las Vegas.

When did you join BD? What would you say fosters your innovative thinking in the workplace?

Three years ago, I joined BD right after graduation from my Ph.D., so I did not have much experience. I was originally hired as a fluidic engineer, mainly because the device I would be working on uses microfluidics, which was the topic I focused on in graduate school.

I was attracted by the job description to develop a new automated platform. We were a small team, so I did both wet lab work and hardware design. I helped with an automated liquid handling instrument for library preparation.

Now, I am continuing to work on our other automated instrument. My primary goal is to enhance processes and efficiency through automation. This approach will significantly reduce the time spent on repetitive tasks, allowing us to focus more on data analysis and other research activities.

How were you first introduced to AI and its applications in health care?

AI’s journey in health care started way back in the 1960s, with early applications being mostly theoretical. By the 1980s, medical imaging advances like CT scans and MRIs spurred interest in using AI for image analysis. The 2000s saw the rise of machine learning and data-driven approaches, expanding AI’s role even further.

More recently, rapid advancements have transformed the field, culminating in breakthroughs like AI-assisted protein structure prediction, which earned the Nobel Prize in chemistry last year. Also, the number of FDA-approved AI-enabled medical devices has skyrocketed from under 50 before 2015 to over 1,000 today.

I took a course in graduate school on machine learning and did a project related to my research. That was probably where my journey started, but I didn’t know all the power it had back then. When ChatGPT launched back at the end of 2022, showcasing its great power, my interest piqued.

For health care applications, my interest grew mostly after I started at BD. In one project, I was pulled in to help with a task that involved manually counting the wells to find a baseline for the analysis process. There are millions of wells in our tiny device, and each well is roughly the thickness of a hair. It was a lot of work.

I learned later that this manual image inspection is the process in our quality control step to report defects in our devices. Therefore, I strived to see if there was anything that could improve efficiency and the mental and physical health of our colleagues. As a result, we started an innovation project that leveraged machine learning techniques to detect these defects.

What are the things that most excite you about bringing AI into health care?

The potential of improving workflow efficiency and the power of AI in revolutionizing research are immense.

In the previous project I mentioned, we applied a machine learning model to automate defect detection in our single-cell capture devices in the use of immunology research. This process was previously done with manual quality control checks, which took a lot of time and had variability.

By integrating AI, we not only accelerated the process but also improved consistency and scalability. Beyond quality control, AI’s ability to uncover patterns in complex datasets opens up new possibilities in diagnostics, personalized medicine, and drug discovery. I believe these technologies will ultimately lead to better health outcomes.

Another project I did as a stretch assignment with the health equity team involved using advanced machine learning (ML) techniques and other key metrics to promote health access. We leveraged county health data and National Institutes of Health (NIH) cancer data to uncover patterns and predictors that contribute to disparities in different communities. Using these insights, we can better support underserved, cancer-vulnerable communities through various outreach programs.

What made you pursue a master’s degree after a Ph.D.? How do you apply knowledge from your degrees to the work that you do at BD?

I pursued my master’s degree to deepen my understanding of AI, and it has significantly enhanced the work I do. I enrolled in courses focused on AI/ML and robotics, which directly support my work to implement an automated platform in our workflow. Additionally, I took foundational courses that helped build a strong base of knowledge. While there is a lot of information that you can self-learn online, earning a degree provides a more structured and systematic approach.

My Ph.D. equipped me with the knowledge I need for the microfluidic devices that we use at BD. Mostly, it taught me how to approach problems and actively learn new things, including soft skills. The tasks at work involve different parties and departments and need to be broken down to find the root cause if something goes wrong.

Similarly, my Ph.D. experiments taught me how to break down the big picture into smaller parts, which is the same as different aims and phases in the workspace. It also taught me resilience; often, experiments don’t work, and hypotheses are wrong, but we actively learn and find new approaches. I think this is important in the workplace.

Do you have any advice for aspiring engineers looking to explore AI and its applications to health care?

Stay curious and proactive. This field is evolving rapidly, and continuous learning is essential. Have an open mind, explore a broad range of technologies, and keep up-to-date on AI innovations and applications in different industries.

Leverage internal resources. If you are in college, attend webinars and explore other resources. For example, Georgia Tech has the Machine Learning Center, which hosts events and weekly webinars. If you are in the industry, attend company events; for example, BD has hosted AI hackathons.

Check out external resources. Consider taking online courses through platforms like Coursera or edX to build foundational knowledge, and stay informed about government initiatives like the NIH’s Bridge2AI, which support research and development at the intersection of AI in health care.

Additionally, understanding regulatory frameworks is vital. Regulations like the FDA’s Good Machine Learning Practices (GMLP) and Europe’s proposed AI Act emphasize safety, transparency, and data quality for high-risk applications, aiming to harmonize ethical and safety practices worldwide and foster responsible innovation in health care AI.

BD is one of the largest global medical technology companies in the world and is advancing the world of health by improving medical discovery, diagnostics, and the delivery of care.

You can explore engineering career opportunities at BD on our website.

Author Dr. Xueying Zhao Dr. Xueying Zhao is an R&D engineer at Becton, Dickinson and Company (BD), where she integrates device innovation with AI-driven health care solutions. After earning a Ph.D. in chemical engineering from the University of Virginia, her passion for solving and streamlining health care problems led her to pursue a second master’s degree in computer science at the Georgia Institute of Technology.

Share Pin 0 Shares