Driving cost out of medical process
If you want to receive access to our newest white paper please provide your email address.
Medical AI is a powerful tool that can do a lot of good. It’s also glamorous: leads to much innovation and attracts a lot of mainstream media.
Digica Solutions would like to keep you up-to-date with information on Digica services. For information on how we will use your data, international transfers of data and your rights, please see Full Privacy Policy details.
Digica was tasked with detecting features in digitised microscopic tissue images. The project was unorthodox due to the scarcity of labelled examples and the target images having resolution in the order of thousands of megapixels.
The project demanded segmentation or detection of features such as cell nuclei, artefacts and microvessels.
We used a small radar system to monitor a baby's breathing pattern.
For every 2000 babies born, one of them will stop breathing in the middle of the night and die. We want to prevent this by creating an automatic, remote breath control device that triggers an alarm when the baby has difficulty breathing.
We created a cross platform library for in-vehicle and remote diagnostic human monitoring.
Using photoplethysmography (PPG) and a standard RGB camera, blood volume changes were detected to measure heart rate variability, respiration rate and blood oxygen saturation.
Driven by computer vision and synthetic data, a deep learning model is incorporated into a mobile application allowing medical personnel to automatically identify medical trays and their contents for use in surgery.
Before surgery, staff take a picture of a surgical tool tray using their smartphone and are notified if instruments are missing.
After surgery, staff are able to determine which instruments have been used or may be missing.
Supporting people with impaired vision by detecting nearby obstructions has been achieved by applying computer vision through a camera integrated into a walking stick.
Our AI-based object detection system recognises surfaces and moving objects based on proximity and potential hazard and communicates this via a haptic controller and Bluetooth earpiece.
The first prototype implementation is being trialled with Blackworld, a blind and partially sighted organisation in Poland