Case Studies

Discover how our innovative solutions have transformed healthcare and technology

Our Proud Partners

R&D Advanced Algorithms & Visualisation

Partner: VeinTech Duration: 9 months

We conducted R&D for innovative algorithms using Python pipelines, with system-based modelling and embedded device compute optimisation.

Key Achievements:

  • Collaboration with Veintech R&D Team to develop software algorithms
  • Partnered with Veintech for implementation of novel Visualisation system
  • Reduced compute time through algorithm optimisation
  • Signal processing pipeline development
VeinTech project image

Feasibility Study for AI-Driven Diagnostics

Partner: lubdub Duration: 2 months

Worked with the LubDub team to conduct a feasibility study for state-of-the-art AI-driven diagnostics.

Key Achievements:

  • Jointly assessed the landscape of AI applications in healthcare
  • Co-developed strategic recommendations for patient outcomes and journey
  • Collaborated on analysis of AI feasibility in cardiac diagnostic processes
LubDub project image

Internal Projects

Real-time AECG Analysis with ML

Duration: 10 months

We developed a live ML inference model with full infrastructure for rapid ECG analysis, including an integrated ML labelling pipeline compliant with ISO standards.

Key Achievements:

  • Achieved 99.7% accuracy in ECG interpretation
  • Reduced analysis time from minutes to under 2 seconds
  • Implemented ISO 13485:2016 compliant labelling pipeline
  • Increased early detection of cardiac anomalies by 42%

ML-Urgency System: Real-time Arrhythmia Detection

Duration: 10 months

Developed a real-time arrhythmia detection system using AWS Sagemaker and Lambda. The system uses a deep learning model trained on the MIT-BIH dataset to detect arrhythmias in ECG data with 98% accuracy.

Key Achievements:

  • Achieved 98% accuracy in real-time arrhythmia detection
  • Deployed on AWS Sagemaker with Lambda integration
  • Integrated with existing ECG data management systems

Clinical Data Labelling GUI

Duration: 6 months

We designed and implemented a user-friendly data labelling GUI for clinical operators, enhancing efficiency and accuracy in medical data processing for large-scale health studies.

Key Achievements:

  • Engineered solution in collaboration with healthcare professionals, ensuring alignment with industry standards and clinical requirements
  • Implemented user-centric design principles to streamline workflow efficiency, resulting in significant improvements in medical data processing productivity
  • Leveraged cloud technology for seamless integration with machine learning algorithms via AWS, enhancing data analysis capabilities and scalability

Our Open Source Projects

Dynamic-ECG: Algorithms for ECG Signal Analysis

Client: Open Source Project Duration: Ongoing

A Python-based library for ECG analysis, including R, P, T wave detection, Poincare analysis, wavelets-based analysis, and several visualisation features! Portability with both short ECG & Long Form Holter data.

Key Achievements:

  • Supports multiple data formats: NumPy, EDF, H5, Apple Watch ECG (CSV export)
  • Advanced wavelet-based ECG analysis
  • Visualisation features for detailed ECG review
Read More on GitHub
Dynamic-ECG project image

SMART BOOT: Assistive Smart Orthopaedic Sensor Device

Duration: 12 months

Constructed Gait-Force frequency Algorithm Development using non-linear differential equation modelling, implemented with Python. Engaged in signal engineering & sensor design, focused on high bandwidth data optimisation.

Key Achievements:

  • Developed a custom gait-force frequency algorithm
  • Optimised sensor performance for high bandwidth data
  • Implemented a biosensor area monitoring system
Read More on GitHub
SMART BOOT project image

CatchAF: Multi-Modal Atrial Fibrillation Detection Model

Client: Research Collaboration Duration: 12 months

An AF detection model that uses Dynamic-ECG for Poincare-Plot generation as input for a Computer Vision model. The model, trained on the IRIDIA-AF dataset, achieved 98% accuracy in detecting AF from ECG data.

Key Achievements:

  • Achieved 98% accuracy in AF detection
  • Integrated Dynamic-ECG for enhanced data input
  • Utilised the IRIDIA-AF dataset for training
Read More on GitHub
CatchAF project image

Ready to Innovate?

Let's discuss how our services can elevate your healthcare technology

Get in Touch