FEATURE: LIEF SCIENCE
numerical values, each representing the position and movement of a part of the body across three dimensions. Once this numerical picture is built, the system needs to be able to make decisions about whether the things it ‘sees’ are normal or not. “We want to know things like whether
the person is moving more slowly now or if there is some asymmetry,” says Fisher, “so we are designing a process that takes all these numbers and maps them somehow onto a decision. It’s possible to use a variety of AI technologies for that, many of which are based on deep learning. We can also present the information directly to the doctors, nurses, and physios so they can assess the situation by themselves.” For Fisher’s team, the next big step is
to develop models that will help the AI make these decisions straightaway, rather than needing time to learn about each individual’s gait and habits. “It would be great to look at a lot of people and build up a model so that, when you install the equipment in someone’s home, it’ll work right off. One of the things we’re looking at now is how much data we need to build that model if it’s possible to do so.” The tools Fisher’s team are developing
could present an alternative to wearables, which are often stigmatised among older people. Wearables also present the challenge of needing to be remembered
www.imveurope.com | @imveurope
and put on each morning. Cost may be an issue for some people, however. “The prototype we’ve made costs
around £500, including the camera and CPU processor. That’s an expensive proposition,” Fisher says. “But on the other hand, a £500 box that helps you stay in your home for another six months and saves you six months of care costs might be a worthwhile investment.”
Instant blood test results – no lab required Sight DX is an Israeli start-up specialising in machine vision for blood analysis. Its Sight OLO diagnostic analyser can provide a complete blood count (CBC) – a standardised suite of blood tests often used as a first-line screening tool, as well as to diagnose and monitor health conditions. But unlike most other providers, the Sight technology doesn’t need to be used in a lab, which means there’s no need for patients and clinicians to wait for the results. “This innovation ultimately makes
fast and accurate diagnostics accessible in a variety of clinical settings – even those without big and expensive central laboratories,” says Sarah Levy of Sight, writing in Forbes magazine. Usefully, the OLO doesn’t require
reagents to be added or maintenance to be done by the tester. It only requires two
“A unit that could be mistaken for a kitchen coffee maker… produces a comprehensive image of the blood sample”
drops of blood in the single-use cartridge. These small, but information-dense droplets, spread out on the surface of the cartridge and are subjected to 1,000 photographs once the cartridge is slotted into the OLO machine – a benchtop-sized unit that could be mistaken at first glance for a kitchen coffee maker. The system’s internal fluorescence microscope produces a comprehensive image of the blood sample that adds up to about 6GB of data. From there, everything is digital. An
array of convolutional neural networks (CNN), physics-based models and feature-based computer vision gets to work identifying and characterising red blood cells, white blood cells and platelets within the sample, and flagging those that show abnormalities. It’s all done in a couple of minutes.
In vivo microscopy We have previously reported on the Virtual Patient project run at the Insigneo
DECEMBER 2023/JANUARY 2024 IMAGING AND MACHINE VISION EUROPE 19
Shutterstock/AHave a nice day Photo
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32
