News RESEARCH & EVENTS From eye to AI:
Energy-efficient vision for intelligent devices
A European consortium led by VTT Technical Research Centre of Finland has developed a machine vision system that mimics how the human eye and brain work — bringing intelligence directly into devices while drastically reducing energy use.
The MISEL project (Multispectral Intelligent Vision System with Embedded Low-Power Neural Computing), funded by the EU’s Horizon 2020 programme, has created neuromorphic circuits that process visual data at the edge.
This allows robots, drones, and smart cameras to interpret their surroundings in real time
without relying on cloud computing or heavy batteries.
“Our aim was to build devices that can see, understand, and act independently,” said Jacek Flak, MISEL project coordinator at VTT. “By replicating the retina and brain’s approach to vision, we can cut power use by hundreds or even thousands of times compared with conventional digital systems.”
The system integrates imaging, processing, memory, and AI algorithms on a single chip. High-speed, high-dynamic-range sensors detect motion and changes — not static
frames — producing compressed, actionable data. Quantum dot sensors extend vision into the infrared, enabling devices to operate in low light or fog.
MISEL’s technology has applications ranging from autonomous drones conducting rescue missions to industrial robots navigating safely among humans.
The co-designed hardware and software platform ensures devices are compact, fast, and energy-efficient, ready for real-world deployment.
Project partners include universities in Finland, Sweden, Germany, Spain, and France, along with Kovilta Oy and AMO GmbH. Their combined expertise in materials science, electronics, and AI pushes machine vision closer to the efficiency of biological systems.
“With these chips, devices can respond to their environment almost as efficiently as a fruit fly — tiny, autonomous, and incredibly fast,” Flak added.
More information online:
ilmt.co/PL/xDXn 66379pr@reply-direct.com
A recombinant laminin matrix from Amsbio has played a key role in a new study from the Stanford University School of Medicine, where researchers used stem-cell-derived blastoid models to explore the earliest stages of human development.
The work [1], published in Nature, demonstrates how blastoids — structures generated from pluripotent stem cells that mimic human blastocysts — can be used to interrogate the influence of human-specific LTR5Hs elements during early embryogenesis. The study offers rare insight into epiblast differentiation and mechanisms shaping the
earliest phases of development.
At the centre of the team’s culture system was iMatrix-511, a purified human recombinant laminin-511 E8 fragment produced in CHO-S cells. Long regarded as a benchmark matrix for maintaining ES and iPS cells under feeder-free conditions, iMatrix-511 supports robust single- cell passaging and delivers stronger adhesion compared with traditional matrices such as full-length laminin, vitronectin, or Matrigel.
The matrix also integrates into Amsbio’s wider stem cell workflow, working alongside the company’s StemFit media and CELLBANKER
The latest news from the science industry by Gwyneth Astles
Stem cell matrix supports breakthrough in human embryo modelling
cryopreservation reagents as part of a complete solution for ES/iPS cell culture.
The Stanford study highlights how reliable, chemically defined matrices can underpin advanced stem-cell-based embryo models — helping researchers probe questions impossible to answer in human embryos.
More information online:
ilmt.co/PL/b8L2
1. A human-specific regulatory mechanism revealed in a pre-implantation model published in Nature
66330pr@reply-direct.com
Cell colonies grown with iMatrix-511. Credit: EPLF-Lutofl Lab
Machine learning untangles complex single-cell data
analyse multiple samples – each with different proportions of cell types, or cell types present in one sample but not another – existing computational methods often falter. Distinct cell populations can be mistakenly merged, masking biological differences.
L-R: António Sousa and Sini Junttila. Credit: University of Turku
Scientists at Turku Bioscience Centre, University of Turku, have developed a machine-learning method to make sense of one of single-cell analysis’ biggest challenges: comparing complex, uneven datasets across samples.
Modern single-cell technologies capture thousands of molecular measurements from individual cells, revealing extraordinary cellular diversity. But when researchers
The Turku team has tackled this using a new algorithm, Coralysis, developed in Professor Laura Elo’s Computational Biomedicine group. Instead of forcing datasets together in one step, Coralysis uses a staged integration approach inspired by assembling a jigsaw puzzle: cells are grouped from simple to increasingly detailed features, progressively refining cell identities with each clustering round.
The result is a tool that can reliably integrate imbalanced datasets and detect subtle shifts in cellular state that conventional methods frequently overlook.
“We wanted a method that uncovers these hidden patterns without collapsing rare or
uneven cell types,” said Associate Professor Sini Junttila, who co-supervised the work.
Coralysis is available as open-source software, and its machine-learning engine can be trained to predict cell identities in entirely new datasets – with built-in confidence estimates. This avoids laborious manual annotation and improves reproducibility across studies.
Lead developer António Sousa described the goal: “Like assembling a puzzle, we start with simple features and build up. Coralysis integrates cells in the same stepwise way.”
Professor Elo added that openly releasing the tool will help accelerate discoveries across the field: “Coralysis gives researchers a new way to navigate cellular complexity.”
The study [1] by Elo’s research group has been published in the scientific journal Nucleic Acids Research.
More information online:
ilmt.co/PL/N484
Illustration highlighting the difference between analysing single-cell data without or with integration. Credit: University of Turku
1. Coralysis enables sensitive identification of imbalanced cell types and states in single-cell data via multi-level integration published in Nucleic Acids Research
66266pr@reply-direct.com
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 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84
