Segmenting and classifying tuberculosis, pneumonia and COVID-19 in chest X-rays
A deep learning pipeline that first isolates the lungs in a chest X-ray, then classifies the image as normal, tuberculosis, pneumonia or COVID-19.
Computer vision
Image and video analysis, from crop disease in a phone photo to vehicle damage in a claim.
Overview
Computer vision turns photos, scans and video into decisions. We build detection, classification and segmentation models that run in the cloud or on a phone, so they work even where connectivity is poor.
What we build
- Detection and classification
- Recognise diseases, defects, damage or objects in images and video.
- Segmentation and measurement
- Outline regions precisely to measure area, size or severity.
- Document vision (OCR)
- Read printed and handwritten forms, IDs and invoices, including Indian scripts.
- Edge deployment
- Compact models that run on smartphones and edge devices without a constant connection.
Use cases by industry
Where computer vision makes the biggest difference for our clients.
Computer vision: case studies & projects
Filter all case studies
Transformer models for coronary artery wall segmentation in intravascular ultrasound
A review of transformer and attention-based deep learning architectures for outlining coronary artery walls in IVUS images, and where they outperform earlier networks.
Crop disease detection and advisory for a farmer producer organisation
Farmers photograph an affected plant; the app identifies the likely disease and gives treatment advice in Marathi, Hindi or English, while the FPO sees outbreaks on a map.
Motor claims intake agent with photo damage assessment
An AI agent that collects claim documents and photos, checks them against the policy, estimates damage from images and prepares the case for a surveyor or adjuster.
How we deliver
Discover
A short workshop to understand the problem, the data and what success looks like.
Prototype
A working proof of concept on a sample of your data, so you can judge results early.
Build
The full solution, tested against your data and integrated with your systems.
Deploy and support
Production deployment on your cloud, with monitoring and continuous improvement.
Questions
Can it work offline in rural areas?
Yes. We compress models to run on the phone and sync results when a connection is available.
How many labelled images are needed?
Often a few hundred per class to start, using pre-trained models. We help plan and run labelling.
More technologies
Explore computer vision for your organisation
Start with a free discovery call. We will tell you honestly what is possible with your data.