Developed ROS2 packages to enable keyboard-controlled actuation of a camera mounted on the rover, providing a complete view during competitions.
Integrated Micro ROS with a Teensy 4.0 microcontroller to process keyboard commands and control servos for precise camera movement.
Collaborated with team members through weekly meetings to ensure seamless integration of hardware and software.

OncoVision is an AI-powered diagnostic tool that uses computer vision to assess whether a visible lump may be cancerous. Built with a convolutional neural network (CNN) trained on medical imaging datasets, the system analyzes images captured by a camera in real time and classifies the lump as likely benign or malignant. It uses preprocessing techniques such as normalization and segmentation to isolate the region of interest, and provides a prediction alongside a confidence score.

Developed in Python using TensorFlow and OpenCV, OncoVision is designed to run on accessible hardware such as a Raspberry Pi with a camera module, making it deployable in remote or low-resource settings. The tool is intended as a non-invasive, early screening support system for healthcare workers or telehealth platforms, helping flag potentially concerning growths for further clinical evaluation. Privacy, fairness across skin tones, and model transparency were key design priorities throughout the project.