inSTEMM Global
60 subscribers
15 photos
113 links
Download Telegram
Unremarkable AI.

Carnegie Mellon University researchers say clinical AI tools should be designed to take tough life-and-death clinical decisions out of the hands of physicians. They suggest that AI might guide decisions best if it were seamlessly embedded in the decision-making routines already used by the clinical team, providing predictions and evaluations on the go.

Read more about their ideas in Proceedings of the CHI Conference on Human Factors in Computing Systems: https://doi.org/10.1145/3290605.3300468

#sciencenews #AI #healthcare #medicine
AI fighting breast cancer.

According to the WHO, breast cancer has recently overtaken lung cancer to become the most common cancer globally. The BreastPathQ Challenge was launched at the SPIE Medical Imaging 2019 conference to support the development of computer-aided diagnosis for assessing breast cancer pathology. 39 teams from 12 countries participated, with 100 new algorithms developed.

The exciting results and ideas produced by the challenge have recently been reported in the Journal of Medical Imaging: https://www.spiedigitallibrary.org/journals/journal-of-medical-imaging/volume-8/issue-03/034501/SPIE-AAPM-NCI-BreastPathQ-challenge--an-image-analysis-challenge/10.1117/1.JMI.8.3.034501.full?SSO=1

#sciencenews #AI #ML #healthcare #medicine
Self-learning robots.

Researchers from AMOLF's Soft Robotic Matter group have shown small, autonomous, self-learning robots can adapt easily to changing circumstances. They connected a group of simple robots in a line, after which each individual robot taught itself to move forward as quickly as possible.

The results are available in PNAS: https://www.pnas.org/content/118/21/e2017015118/tab-article-info

#sciencenews #AI #robots
Artificial muscles.

Universidad Carlos III de Madrid researchers offer guidance on the design of magneto-active structural systems that can be applied to stimulate wound healing and artificially replicate muscle tissues. They describe their method as creating an ‘athletic track for cells’.

Two articles on their work have been published recently in Composites Part B: Engineering and International Journal of Solids and Structures: https://doi.org/10.1016/j.compositesb.2021.108796 https://doi.org/10.1016/j.ijsolstr.2020.10.028

#sciencenews #AI #bioengineering
A graphene key for computing.

Current silicon technology exploits microscopic differences between computing components to create secure keys, but AI techniques can be used to predict defects and gain access to data. Penn State researchers have designed a way to make the encrypted keys harder to crack using graphene.

The results are presented in Nature Electronics: https://www.nature.com/articles/s41928-021-00569-x

#sciencenews #AI #computing #graphene
AI-powered‌ ‌microscopes.‌ ‌

Light‌ ‌field‌ ‌microscopy‌ ‌allows‌ ‌the‌ ‌neuronal‌ ‌signals‌ ‌in‌ ‌the‌ ‌brain‌ ‌to‌ ‌be‌ ‌imaged‌ ‌in‌ ‌real‌ ‌time,‌ ‌but‌ ‌the‌ ‌images‌ ‌are‌ ‌often‌ ‌lacking‌ ‌quality‌ ‌and‌ ‌take‌ ‌a‌ ‌long‌ ‌time‌ ‌to‌ ‌process‌ ‌for‌ ‌visualisation.‌ ‌European‌ ‌Molecular‌ ‌Biology‌ ‌Laboratory‌ ‌scientists‌ ‌are‌ ‌using‌ ‌artificial‌ ‌intelligence‌ ‌to‌ ‌boost‌ ‌the‌ ‌image‌ ‌processing‌ ‌speeds‌ ‌from‌ ‌days‌ ‌to‌ ‌seconds.‌ ‌

Learn‌ ‌about‌ ‌their‌ ‌technique‌ ‌in‌ ‌Nature‌ ‌Methods:‌ ‌
https://www.nature.com/articles/s41592-021-01136-0‌

#sciencenews #AI #science #microscopy
On the brink of chaos.

Scientists at the University of Sydney and Japan's National Institute for Material Science have discovered that an artificial network of nanowires can be tuned to respond in a brain-like way to electrical stimuli. By keeping the network of nanowires in a chaotic, brain-like state optimized its performance.

Their insights are published in Nature Communications: http://dx.doi.org/10.1038/s41467-021-24260-z
#sciencenews #nano #AI
AI in 3D printing.

Additive manufacturing allows on-demand production. However, the performance of the final object is hard to predict. A team at the University of Texas has shown that neural networks can be used to better understand the processes.

The study is published in the journal Computational Methods: https://www.sciencedirect.com/science/article/abs/pii/S0045782521002474?via%3Dihub

#sciencenews #AI #3dprinting