We recently had the opportunity to sit down with Dmytro Zaharnytskyi, an AI and Machine Learning Engineer with a compelling vision for the future of biodefense. Dmytro, who has extensive experience in AI-driven solutions for both public health and military applications, is dedicated to furthering his work on developing an AI-driven biological detection and defense software system. We spoke about his background, motivations, and vision for the future.
Dmytro, could you start by telling us a bit about your background and your journey into AI and biodefense?
Of course! I started my journey in technology and computer sciences with a bachelor’s degree in cybersecurity, which helped me understand the challenges of protecting critical infrastructures. I then pursued a master’s in AI systems, which expanded my horizons to machine learning and its vast applications. Over the last five years, I’ve worked on various projects involving machine learning operations, biotech, and anomaly detection, including applications for military safety and public health. One of the pivotal experiences for me was working on computer vision models for battlefield object detection, which really opened my eyes to how AI could be used to save lives in high-risk environments.
What led you to focus specifically on the biodefense sector?
My motivation stems from the realization that biological threats, whether they are pandemics or bioterrorism events, represent a critical challenge to our society. The COVID-19 pandemic highlighted how quickly a biological threat can overwhelm healthcare systems worldwide. My goal is to create an AI-driven software system that can provide early detection of these threats by analyzing deviations in environmental and hospital data. By catching anomalies early, we can provide a faster response, which could potentially save thousands, if not millions, of lives.
Can you explain in more detail what your proposed AI-driven biological detection and defense software system would do?
Absolutely. The idea is to create a comprehensive detection and response system leveraging advanced AI techniques to monitor anomalies. We would integrate environmental sensors and analyze hospital data streams in real time. For instance, if a hospital network begins reporting unusual symptoms or an environmental sensor detects a chemical marker that shouldn’t be there, the AI can correlate these data points to identify emerging threats. The system would use anomaly detection algorithms to provide a clear picture of potential risks, which could indicate anything from a natural disease outbreak to a bioterrorism event. Of course, it is the end goal, for starters it will be enough to find deviations in data that are already there, area specific air quality, viral activity in wastewater, data from in-house devices like Alexa, hospital attendance numbers, crowd sizes, car accident amounts, etc.
How would this system differ from existing public health monitoring systems?
Current public health monitoring systems tend to rely heavily on manual reporting and are often reactive rather than proactive. What I envision is a shift towards real-time data collection and AI-driven anomaly detection. By automating the data analysis process, we could achieve faster identification of unusual patterns and initiate responses much more quickly. Imagine a network that is constantly analyzing environmental, medical, and social data, and able to send alerts as soon as something seems out of the ordinary—that’s the type of proactive defense mechanism I want to build.
You mentioned using anomaly detection techniques for identifying biological threats. Could you tell us more about your experience with these techniques?
Throughout my career, I have focused on anomaly detection and its applications across various fields. At Neurotrack, I was leading the team, and we had created an EEG-based BCI system. This project deepened my understanding of how to analyze brain signals and most importantly deviations in them to identify emotional states, contributing to advancements in neurotechnology.
In addition to my work at Neurotrack, I have gained experience in anomaly detection during my time in the Ukrainian military project. While it was mostly centered around computer vision, we also employed anomaly detection techniques to identify unknown or suspicious objects on battlefields, as sometimes unexpected objects could appear and the model should mark them also.
The inspiration for my interest in biotech actually began in my adulthood, specifically in the ninth grade. During a nationwide ecology project, my team and I analyzed the induction of chlorophyll fluorescence in different plants. This research aimed to understand the levels of pollution in our environment and to explore which plants could help mitigate these issues—long before artificial intelligence became mainstream. This early experience sparked my passion for using data analysis to identify and address real-world problems.
How do you see AI evolving in terms of its role in national and global security?
AI has the potential to transform national and global security, especially in areas like biodefense, where speed and accuracy are crucial. We’re at a point now where we can use machine learning not just to analyze data, but to derive insights and make decisions in real-time. I believe AI will play an increasingly vital role in preventing crises before they happen, from natural pandemics to bioterrorist attacks. It’s about leveraging data and technology to create a layer of defense that’s proactive rather than reactive.
What are some of the biggest challenges you face in making this vision a reality?
One of the biggest challenges is data integration. The kind of system I’m proposing needs access to various data sources—hospital records, environmental sensors, and other public health databases—and integrating these different types of data in a coherent and secure manner is a complex task. Another challenge is ensuring privacy. Using medical data always raises privacy concerns, so developing a system that can extract useful information without compromising individual privacy is key. Lastly, there’s the challenge of gaining trust and collaboration across different institutions—governments, hospitals, and research bodies all need to work together for a solution like this to be effective.
Finally, what drives you to keep pushing forward in this field?
I think the most motivating factor for me is the potential to make a real difference. AI is a powerful tool, and if we can use it to protect people from biological threats, we have a moral obligation to do so. My work in biodefense and my experience in AI have shown me that technology when used responsibly, can save lives. That’s what drives me every day—knowing that the work I do might help prevent the next pandemic or mitigate the impact of a biological attack. I lost my father in the war in Ukraine, and this experience has given me a profound understanding of the value of human life. While I cannot stop wars around the world, I do have the opportunity to use my skills to save lives and make a meaningful impact.
