Documents

AI-Powered Gait Analysis

Run inference on the LSTM model deployed to Google Cloud Vertex AI using 50 samples of sensor data

LSTM Time-Series Forecasting Model

What This Model Does

This LSTM neural network analyzes 50 consecutive timesteps of sensor data and predicts the next timestep's values for all 15 sensor features. It's a time-series forecasting model, not a classification model.

R² Score
96.99%
Model accuracy
MAE
0.103
Mean Absolute Error
RMSE
0.174
Root Mean Square Error

How to Use

  1. 1
    Upload or paste 50 samples of sensor data (CSV or JSON format)
  2. 2
    Each sample must contain all 15 sensor features (HR, SpO2, AccX, AccY, AccZ, etc.)
  3. 3
    Click “Run Inference” to predict the next timestep
  4. 4
    View predicted sensor values for the immediate future

Required Sensor Features

15 features required in exact order: HR, SpO2, AccX, AccY, AccZ, GyroX, GyroY, GyroZ, Aroll, Apitch, Groll, Gpitch, Gy, Combroll, Combpitch

LSTM Model Inference

Run predictions on the LSTM model deployed to Google Cloud Vertex AI

Checking...

Input Data

Provide 50 samples with 15 features each