What if we could take something as simple - and seemingly ordinary - as a urine sample, measure thousands of chemical signals within it, and use artificial intelligence to recognize the metabolic fingerprint of disease?
And what if, instead of looking for one specific biomarker, we could train an algorithm to recognize the complex pattern created by an entire disease state?
That is the premise behind Luventix ( https://www.luventix.com/ ), a technology company developing AI and machine-learning diagnostic models based on quantitative analysis of urine.
Today's guest, Dejan Nenov, comes to this problem from a somewhat unusual direction.
He's not a traditional biotech entrepreneur.
Dejan is a computer engineer, technology entrepreneur, investor and longtime software executive who has spent more than three decades building companies across fields ranging from telecommunications and cybersecurity to enterprise software, consumer electronics and data architecture.
Dejan has been involved with companies that have collectively returned more than $3.5 billion to investors - and he's now turned his attention to one of the most complicated systems imaginable:
the human body.
At Luventix, the idea is to use gas chromatography to generate a quantitative representation of the metabolic state contained within a urine sample, and then use machine learning to identify patterns associated with disease, treatment response, and potentially treatment-related toxicity.
Today on Progress, Potential, and Possibilities, we're going to explore whether the next generation of medical diagnostics may be less about finding a single molecule - and more about teaching machines to recognize the incredibly complex biological patterns hidden in our own chemistry.
Podden och tillhörande omslagsbild på den här sidan tillhör
Ira Pastor. Innehållet i podden är skapat av Ira Pastor och inte av,
eller tillsammans med, Poddtoppen.
