Inspiration
Chronic wounds affect millions worldwide, especially patients with diabetes, limited mobility, or poor access to advanced healthcare. Many treatments are expensive, slow, or inaccessible outside clinical settings.
We were inspired to explore how emerging biomedical technologies like cold atmospheric plasma could be combined with robotics and automation to make wound treatment faster, safer, and more accessible.
Our goal was to make an analysis unit that could integrate with any robotic arm to ensure cold plasma application safety and make semi-autonomous or full-autonomous application viable.
What it does
The Multimodal Plasma Wound Therapy Robot analysis unit delivers controlled cold plasma therapy to a wound site using an automated robotic platform. It is designed to assist in sterilization, promote tissue regeneration, and reduce infection risk. The system integrates multiple modalities including sensing, positioning, and plasma delivery to improve precision and consistency compared to manual application.
How we built it
We designed a robotic platform capable of controlled movement over a target surface, paired with a plasma generation system calibrated for safe therapeutic exposure. We developed the system architecture to include hardware control, safety constraints, and software logic for repeatable treatment patterns. Prototyping involved iterative testing of mechanical design, electrical components, and control behavior to balance effectiveness with safety.
Our system captures depth maps of a patient’s wound area using an Intel RealSense camera and converts these depth images into 3D point clouds for geometric analysis. Each depth pixel represents the distance from the camera to the surface, and using the camera’s intrinsic parameters, these pixels are back-projected into 3D space. This process reconstructs the wound surface as a dense set of 3D points, enabling accurate measurement of wound depth, area, and surface morphology.
A thermal camera is used to detect anomalies and ensure that the patient's wound area remains within the safe operating temperature, which provides feedback to our path generation algorithm.
We've developed a path generation algorithm to generate G-Code data, which can provide tool path information to any robotic actuator. This takes in the 3D point cloud data and uses AI to detect the would area to generate a path for the plasma jet to uniformly distribute dose to the patient's wound area.
Challenges we ran into
One of the biggest challenges was working within safety constraints while still demonstrating meaningful functionality. Plasma systems require careful tuning, and we had to ensure the design emphasized control and predictability. Integrating hardware and software smoothly also required multiple design iterations. Time constraints pushed us to prioritize core functionality over advanced features, which forced thoughtful tradeoffs.
Accomplishments that we're proud of
We successfully built a functional prototype that demonstrates the feasibility of automated plasma-based wound therapy. We created a cohesive system rather than disconnected components, showing real-world potential beyond theory. We are especially proud of designing with safety, accessibility, and practical medical impact in mind rather than focusing only on technical novelty.
We plan to implement real-time sensing and feedback, improving precision using computer vision, and exploring more compact designs for portability.
What we learned
We gained hands-on experience with interdisciplinary engineering across robotics, biomedical design, and systems integration. We learned how critical iteration is when building physical systems, and how design constraints can sharpen rather than limit innovation. We also developed a stronger understanding of how emerging medical technologies can be translated into practical tools.
What's next for Multimodal Plasma Wound Therapy Robot
We want to refine the system toward clinical viability and continue researching how automation can expand access to advanced medical treatments in both clinical and home environments.
Log in or sign up for Devpost to join the conversation.