Engineer to Athlete
A Flow Assurance Engineer Runs Into the World of Data Science
Introduction
I first met Denis Akhiyarov 15+ years ago at Multiphase Solutions (MSi). We worked together, co-authored technical papers (“Minimum Flow Rate To Unload Gas Wells: A New Dynamic Modeling Pipe Flow of High-Viscosity Production Fluids during Transient Operations”, St. Petersburg, Russia), and even attempted to sail a cardboard boat together in Mexico. (This is one of my favorite work pictures ever, what a team!)
Denis / I went our separate ways, me via Assured Flow Solutions and now Pontem, while Denis has gone on to do several things, including roles in Machine Learning with TotalEnergies and Google. Currently, Denis and his partner Anton Galvas (another oilfield survivor) have started Aikynetix, which is a focused on sports performance.
With a shared vision for leveraging data to provide practical, real help to people, Pontem Performance is collaborating with Aikynetix in pursuit of opportunities that bridge the gap between data and discipline in the field of athletics. We sat down with Denis and Anton for some insights on Aikynetix and their vision for how to better leverage data for practical insights.
Interview
Why did you start Aikynetix?
We started Aikynetix with a vision to revolutionize the way athletes and individuals optimize their performance and prevent injuries. We saw a significant gap in the market for a comprehensive solution that combines advanced computer vision technology, biomechanics expertise, and data science to provide actionable insights and personalized feedback. Our passion for helping athletes reach their full potential and our belief in the power of technology to enhance human performance were the driving forces behind starting Aikynetix.
What makes Aikynetix unique?
Aikynetix stands out due to our unique approach to sports performance analysis and injury prevention. We leverage biomechanics, cutting-edge computer vision algorithms and machine learning techniques to analyze athletes' movements in real-time. Our platform provides detailed biomechanical assessments, which can help to identify potential areas of improvement. We are confident, that with the help of our advisors, customers, and product champions will be able to provide personalized recommendations for training, technique refinement and injury avoidance. What sets us apart is our ability to seamlessly integrate domain knowledge, such as biomechanics, with advanced data science to deliver actionable insights in an accessible and user-friendly manner.
How did you pivot from running to other sports, as well as potential applications outside of athletics?
Throughout our journey, we have continuously expanded our horizons, applying our technology and expertise to various domains. It all began with a focus on running, where we developed a deep understanding of human movement patterns and became aware of the alarming rate of daily injuries among runners. As we demonstrated the capabilities of our technology to prospective investors and customers, there was a demand to explore its application in other sports such as golf, baseball, soccer, and more. This led us to realize the versatility of our technology and its potential in unexpected areas. An exciting turning point came when we showcased our capabilities to a commercial security contact. We demonstrated how our advanced tracking of human motion could activate otherwise "dumb" cameras, providing enhanced security solutions. This successful demonstration opened up an opportunity for us to deploy our technology in a convenience store, where we efficiently detected and reported on a high-value item loss prevention incident within a month. Simultaneously, we observed a growing number of early adopters subscribing from the field of physical therapy and rehabilitation hospitals. This presented a unique challenge compared to working with athletes, as our focus shifted from running to walking and monitoring patients outside of the treadmill environment. Addressing under-striding and muscle weakness issues among rehab patients became essential in their recovery process. Furthermore, we had the privilege of demonstrating our technology to NASA, exploring its potential in space-related activities. This opportunity served as a testament to our adaptability and ability to apply our technology in diverse contexts, further fueling our commitment to innovation and our desire to make a positive impact beyond a single domain. We take great pride in the journey we have embarked upon, and each step has strengthened our belief in the transformative power of our technology. By leveraging our expertise in biomechanics, data science, and computer vision, we are well-positioned to continue making a difference in various fields, enhancing performance, and improving outcomes.
What is the future of computer vision (sports and beyond)?
The future of computer vision holds immense potential across various domains, including sports and beyond. In the sports industry specifically, computer vision technology will continue to play a pivotal role in athlete performance analysis, injury prevention, and game strategy optimization. We anticipate significant advancements in several areas. Firstly, real-time tracking capabilities will become even more precise and comprehensive. This means that coaches, trainers, and analysts will have access to more detailed and accurate data on athlete movements, allowing for more informed decision-making and targeted improvements in performance. Additionally, we foresee advancements in enhanced object recognition with a single view camera. This means that computer vision systems will be able to identify and analyze key elements of the game or activity with greater accuracy and efficiency. This will lead to more sophisticated insights and actionable information for athletes, coaches, and teams. Moreover, the integration of physics and analytics will be further deepened. Computer vision algorithms will be able to incorporate physics-based models and simulations, providing a deeper understanding of the dynamics and mechanics of sports movements. This integration will enable more precise biomechanical analysis, injury risk assessment, and personalized training programs. Another area of development will be the sustainability of algorithms to occlusion problems. Occlusion occurs when objects or athletes are partially or completely hidden from view, posing challenges for computer vision systems. We anticipate significant progress in developing algorithms that can effectively handle occlusion, ensuring consistent and reliable tracking and analysis even in complex real-world scenarios. Looking beyond the realm of sports, computer vision will continue to revolutionize industries such as healthcare, autonomous vehicles, augmented reality, surveillance, and many others. As computer vision algorithms become more sophisticated and accessible, we envision a future where this technology becomes an integral part of our daily lives, enhancing our health, safety, and overall well-being. The potential for computer vision is vast, and we are committed to staying at the forefront of these advancements, leveraging our expertise and innovation to drive positive change across multiple domains.
How critical is the combination of machine learning and domain knowledge (i.e. biomechanics)?
The combination of machine learning and domain knowledge, like biomechanics, is essential for unlocking the full potential of sports performance analysis and injury prevention. While machine learning provides the tools to process large volumes of data, relying solely on traditional "computer vision and AI only" applications can present challenges and sustainability problems. To address these challenges, our approach at Aikynetix goes beyond the black box. We strive to demystify the technology by providing actionable human motion insights. By integrating domain knowledge, such as biomechanics, we gain a more comprehensive understanding of the underlying mechanics and intricacies of human movement. This allows us to deliver more accurate and actionable insights, empowering athletes, coaches, and medical professionals with personalized feedback and guidance. By combining machine learning and domain knowledge, we aim to provide a transparent and effective solution that enhances human performance analysis. Our focus on demystifying the black box and offering granular motion analysis sets us apart and ensures sustainability in delivering valuable results.
Wrapping up
At Pontem, we are excited about our collaboration with Aikynetix as we both seek to leverage data and analytics to help performance at all levels. In the coming months we will be diving further into our new partnership, discussing specific case studies we have worked on together utilizing this shared vision.
For more insights on Aikynetix, check out their website:
And for more information on Pontem Performance, be sure to check out our website as well:
www.pontemanalytics.com/performance
