Heartbeat to Visuals

Responsibilities: Signal Capture with Arduino, Particles simulation, Data Pocessing, Data Visualization

Overview

This project was inspired by the challenge of making invisible physiological signals—like the heartbeat—something people can directly perceive. Instead of presenting data as numbers or charts, I translate heartbeat into real-time visuals and sound, making scientific information more intuitive and engaging.

Goals

• Accessible Science Communication. I transform physiological data into audiovisual experiences to make scientific concepts tangible through art.

• Human-Computer Interaction. I explored heartbeat as a new kind of interface, where the body itself drives interaction with visuals and sound.

Heartbeat Capture

A pulse sensor connected through an Arduino reads the signal, processes BPM and IBI in Arduino IDE, and streams them to the computer. The values are sent over USB as simple data that TouchDesigner can use in real time.

In greater details:

• A Pulse Sensor works by shining a green light onto a thin part of your body. When the light travels through your skin, some of it gets absorbed by your blood, and the rest bounces back. The Pulse Sensor has a special part called a photodetector that measures how much light returns.

• The sensor is powered through the development board. It connects to it with a 5V power input, Ground pin, and other analog input pin that carries the signal.
• The board connects to the laptop through HDMI, where in Arduino IDE I compile and run the code that processes the signal data and allows the heartbeat sensor to communicate with TouchDesigner. I use the Pulse Sensor library. Inside the Arduino loop, the sensor is updated in real time, and then two values are calculated: the beats per minute, or BPM, and the inter-beat interval, or IBI, which is the time between individual heartbeats. These numbers are then sent over USB to the computer in a simple CSV (Comma-Separated Values) format, so TouchDesigner can read them easily. The short delay at the end keeps the updates running smoothly at about 50 times per second.
• The Serial node inside TouchDesigner receives the data from the specified port and puts it into a table.

Data Processing & Visualization

In TouchDesigner, the ParticleGPU node drives the visuals. It spawns particles, applies forces, and maps heartbeat data onto movement, color, and audio.

• BPM changes color, force size, birth rate, and sound tempo.
• IBI controls particle speed.
• At higher BPM, particles accelerate, more are generated, spread wider, and audio and visuals' color mimic the intensity of a racing pulse.
• Post-processing adds bloom, text for BPM/IBI, and a feedback loop that creates depth by layering frames with decreasing opacity.

Custom Velocity Lookup Table

One highlight is the custom velocity lookup table. This defines how turbulence changes as particles speed up:

• At low velocity, turbulence is subtle, keeping particles close to the source.
• At medium velocity, turbulence increases, creating branching, neuron-like flows.
• At high velocity, turbulence becomes chaotic and flickering, giving the structure its energetic quality.
• This gradual interpolation is what produces the system's distinctive organic motion.

Relevance

This work is applicable in education, healthcare, and immersive installations. In education, the system transforms abstract physiological data into visuals and sounds that students can intuitively grasp. In healthcare, the approach can support biofeedback and mindfulness by giving patients immediate, sensory ways to monitor and regulate their own physiological signals. In immersive art, it creates interactive environments where audiences can literally “see and hear” their bodies in real time, fostering personal connection and curiosity.

Supports accessibility. Many audiences, including children, neurodiverse learners, and people with low literacy or vision impairments, can struggle with numbers, graphs, or text. My approach offers non-numerical, non-textual, representation of data, which reaches broader and more diverse audiences.

Future Directions

As I continue my research I plan to firstly, expand the physiological inputs using the Arduino and proximity sensor, depth camera to track body movements hand gestures. Secondly, I want to conduct studies among students and professors to measure understanding, reflection, and engagement. This will help me assess the real impact my project can have in different sectors. Finally, I want to encourage cross-disciplinary collaborations in the field of arts and science. I would like to work with scientists and artists to refine mappings between data and expression.

Challenges and learning outcomes

One of the main challenges in this project was achieving a stronger sense of depth in the particle system.

Initially, the particles generated by the Particle GPU appeared too flat, existing on a single visual layer. I wanted the structure to feel more three-dimensional and volumetric without losing the delicate branching form I had already built. Increasing the birth rate added density but didn't create the illusion of space I was aiming for. The solution was to introduce a feedback loop that layers each frame with slightly reduced opacity—about a 10% decrease per frame—creating subtle visual trails that accumulate over time. This approach gave the particles a soft, immersive depth effect while preserving the original geometry and overall flow of the system.