Process

How FLEXA works

From sensing to actuation in a single closed loop — through embedded boards, the cloud, machine-learning inference, and a clinician-facing dashboard.

End-to-end flow

From the patient's hand to the cloud and back.

01

Capture finger motion and vitals

Five flex sensors on the healthy-hand glove produce analog resistance proportional to bend angle. A heart-rate sensor reads continuously, and an RGB camera captures the patient's face for landmark extraction.

Flex sensors ×5MAX30100RGB camera
02

Convert and bridge through ESP32

Dual-core ESP32 microcontrollers sample the ADCs, convert raw values to angular degrees, run safety checks, and stream data over Wi-Fi to the cloud — all while controlling MOSFET-driven vacuum pumps on the actuation side.

ESP32 240 MHzADS1115 16-bitMOSFET drivers
03

Sync through Firebase Realtime Database

Sensor frames, prescriptions, and device state synchronize through Firebase at 10 Hz with sub-200 ms latency. Patient gloves poll for active prescriptions; clinician dashboards subscribe to live progress feeds.

Firebase RTDB10 Hz polling< 200 ms latency
04

Analyze with machine learning

A Flask API on AWS EC2 receives grip-strength frames and runs a TensorFlow neural network that places each patient on a five-stage recovery scale. NeuroSpeed runs ensemble emotion recognition (DeepFace + FER CNN) at 4–10 fps.

TensorFlowDeepFaceMediaPipeFlask · AWS EC2
05

Actuate the paralyzed hand

Five vacuum-driven soft actuators replicate the prescribed bend angles independently per finger, with force kept below 5 N for clinical safety. Tiered HR rules and emotion signals can slow or pause therapy in real time.

Vacuum micro-pumps< 5 N forceHIGH / LOW / STOP
06

Visualize on the clinician dashboard

Physiotherapists review session history, longitudinal grip recovery trajectories, emotion timelines, and adjust the next prescription — all from a React web application.

ReactFastAPI WebSocket

Four-tier architecture

Layer · 01

Sensor & acquisition

Flex sensors, FSR402 ×6, MAX30100 HR, ADS1115 ADCs, RGB camera.

Layer · 02

Embedded processing

ESP32 dual-core boards: ADC sampling, actuator control, Wi-Fi bridge.

Layer · 03

Cloud backend

Firebase RTDB for state sync, Flask + AWS EC2 for ML, FastAPI WebSocket for streams.

Layer · 04

Presentation

React clinician dashboard plus mobile companion for the patient.