regenRx

Inspiration

The inspiration behind regenRx comes from the critical inefficiencies in modern drug discovery and clinical trials. Nearly 90% of drug candidates fail during clinical testing, majority due to limitations in dosage, safety, or efficacy. Alongside that, antibiotic resistance and other time-based pushes highlight the urgent need for faster and more intelligent drug development methods.

What it does

regenRx is an AI-powered drug discovery and validation platform that integrates computational modeling with microfluidic hardware to accelerate pharmaceutical research. The system identifies promising drug candidates using machine learning, simulates molecular interactions through in-silico modeling, and validates results using organ-on-a-chip technology. It creates a closed-loop pipeline where AI predictions are continuously tested and refined using real-time biological feedback.

How we built it

We built regenRx using a multi-layered architecture combining AI, simulation, and hardware integration. The backend AI system uses graph neural networks and natural language processing to analyze drug databases, genomic data, and clinical research papers. Molecular docking simulations are used to predict binding affinity between drug compounds and target proteins. We then integrated this system with a microfluidic chip platform that allows automated testing of drug concentrations on tissue cultures. A central software hub coordinates data flow between AI models and physical hardware.

Challenges we ran into

One of our biggest challenges we ran into was our AutoCAD file completely deleting, twice. We worked through all of these hurdles as a team that began as strangers. With one member joining late, the teams worked to catch her up and help her understand the complex ideas of AI modeling in drug testing and split off each team member to work on a different component of developing our idea.

Accomplishments that we're proud of

We are proud of successfully designing a full end-to-end pipeline that connects AI-driven drug discovery with real-world biological validation. Our system demonstrates the feasibility of closing the loop between computational predictions and lab-based experimentation. We also developed a scalable architecture that can be extended to multiple disease models and drug classes.

What we learned

Through building regenRx, we learned how critical system integration is when combining AI with physical biology systems. We gained experience in working with highly scientific models that we were not previously familiar with, and microfluidic system design in AutoCad. Most importantly, we learned how interdisciplinary collaboration between computer science, biology, and engineering is essential for meaningful innovation in healthcare.

What's next for regenRx

In the future, predicted drug formulations could be sent directly from regenRx to a connected lab platform, closing the gap between digital simulations and real-world testing. Automated systems could precisely mix and deliver AI-designed drug concentrations into Organ-on-a-Chip models, reducing manual handling and improving accuracy. Once a drug shows successful results in lab validation, the system could also use that data to simulate and optimize how the drug would be produced at a large manufacturing scale, helping speed up the transition from research to production.