Technology

The sensing methods our instruments and electrodes are built on

Core Technologies

Cutting-edge sensing methodologies

Single-Entity Electrochemistry

Ring ultra-microelectrode (UME) technology enabling precise detection and sizing of individual micro- and nanoparticles.

Key Features:

  • • Single-particle resolution
  • • Size range: 300 nm - 5 μm
  • • Current-blockade detection
  • • Label-free measurement

Patent pending: "Electrochemical affinity biosensor for label-free detection using particle collision"

View Poster

MIP-Functionalized Electrodes

Molecularly Imprinted Polymers (MIPs) provide affordable, stable artificial recognition elements for selective molecule detection.

Applications:

  • • PFAS environmental monitoring
  • • Cortisol stress detection
  • • Sweat biomarker analysis
  • • Drug screening

Integrated with our €499 point-of-need potentiostat for affordable dev-kits

View Poster

Machine Learning & Data Analysis

Deep learning applications for biosensor signal processing

StepReaderCNN-MVP

CNN-Based Electrochemical Signal Classifier

Open Source

Deep learning framework for analyzing electrochemical sensor signals. The system classifies particles by size (1μm, 2μm, 3μm) using collision signal data from single-entity electrochemistry experiments, achieving 80% validation accuracy.

🧠 Technologies

  • • PyTorch 2.9.0 (3 CNN architectures)
  • • Streamlit web interface
  • • FastAPI backend
  • • NumPy, Pandas, SciPy
  • • TensorBoard visualization

🎯 Key Features

  • • 42 real CSV datasets (99K-153K points)
  • • Synthetic signal generation
  • • ResNet1D, SimpleCNN, MultiScale models
  • • 104.1 inferences/second
  • • Interactive data exploration

Performance: ResNet1D achieves 80% validation accuracy with ~29 second training time and 9.61ms inference latency

License: MIT License - Open source for academic research

Live Demo View on GitHub