Elonosia AI
prototype/ncair / nitda // ai & embedded systems // 2022
C++ESP32TensorFlowAltium PCB Design3D PrintingPythonCNN
the product
Elonosia AI was an NCAIR-incubated startup building automated malaria detection from blood smear microscopy images. The goal was a device that could sit in a rural clinic and diagnose malaria without a trained microscopist.
what i built
I designed and built the hardware prototype -- an automated microscopy capture device using ESP32/ESP8266 microcontrollers, custom PCBs designed in Altium, and 3D-printed enclosures. The device automated slide positioning and image capture, feeding frames to a TensorFlow CNN model trained on blood smear datasets.
impact
The prototype demonstrated that automated capture and ML-based diagnosis was viable at low cost. I also contributed to training and validating the CNN models for blood smear classification.