- Статьи
- Science and technology
- Technologically: leather from the laboratory and the "heavenly tractor"
Technologically: leather from the laboratory and the "heavenly tractor"
Russian scientists have created a neural network that will detect signs of Parkinson's disease from EEG images, tested the country's first "sky tractor" and developed medical bandages, the use of which reduces blood loss in complex wounds by four times. The first computer virus using AI has been found online. And Australian researchers have grown a skin organoid that completely mimics human skin from reprogrammed stem cells. About these scientific discoveries — in the weekly column "Izvestia".
New nanomaterial will reduce blood loss
Russian scientists from the University of MISIS have developed a new nanomaterial for medical dressings, the use of which reduces blood loss in complex wounds by four times compared with conventional dressings.
— Researchers at our university have developed a bandage designed primarily for severe wounds in conditions of limited medical care. It is made of nanofibre material, the thinnest membrane, the fibers of which are hundreds of times thinner than a human hair. The innovative patch, which has passed clinical trials, is able to reduce blood loss, accelerate wound healing, and destroy a wide range of bacterial strains. The bandage will also be effective in the treatment of diabetic ulcers and complex burns," said Alevtina Chernikova, Rector of NUST MISIS.
Tests have shown that the new bandage reduced blood loss by 4.6 times compared to conventional means. Scientists from Lomonosov Moscow State University and the State Scientific Center for Applied Microbiology and Biotechnology conducted a study of the bactericidal and fungicidal properties of the material, noting its high effectiveness against antibiotic—resistant strains of superbugs - Staphylococcus, Pseudomonas aeruginosa and E. coli, enterococcus and Candida auris fungus.
Neural network will detect signs of Parkinson's disease by EEG
Scientists at Sechenov University have found a way to automate the early diagnosis of Parkinson's disease and make it more accessible to patients. They have developed a neural network to verify the features of the electrical activity of the brain. In the future, it is planned to create the first digital service for early diagnosis of Parkinson's disease based on EEG data. With its help, doctors will be able to quickly and accurately make a preliminary diagnosis and prescribe studies to confirm it.
— Currently, EEG is not used to diagnose this disease. However, there are scientific publications by Russian and foreign researchers devoted to the peculiarities of the electrical activity of the brain on electroencephalography of patients with Parkinson's disease and their comparison with the EEG of healthy people. We set ourselves the task of finding out whether it is possible to differentiate this data using a neural network. As it turned out, this is quite realistic," Ekaterina Vakhromeeva, the author of the project, a graduate of the Master's degree in Information Systems and Technologies at Sechenov University, told Izvestia.
For this work, the scientists used an open foreign dataset containing anonymized EEG data from patients with Parkinson's disease and healthy participants, manually tagged by a neurologist. The researchers divided the data into samples for training and testing the neural network. The model was trained on the first sample, and on the second sample, which she "saw" for the first time, her ability to detect the EEG of healthy and sick patients was tested. As a result, it turned out that the neural network can recognize frequency anomalies of electroencephalograms and identify patients with Parkinson's disease with an accuracy of 97%.
Unmanned "flying tractor"
Specialists of Kazan State Agrarian University have started testing a large agricultural unmanned complex "Heavenly Tractor".
The experiment is conducted at the university's experimental sites. The 150 kg machine is capable of spraying fertilizers and plant protection products on up to 600 hectares of crops per shift. According to the creators, its performance is 8-10 times higher than the capabilities of traditional ground-based sprayers when working on similar areas with difficult terrain. This is the only Russian drone with such capabilities and characteristics.
— The main advantage of drones is the accuracy of application and independence from the soil condition. You can work immediately after the rain, when the ground equipment gets bogged down. But the technology requires highly qualified operators and expensive maintenance," said Rais Sabirov, Associate Professor of the Department of Operation and Repair of Machines at Kazan State Agrarian University.
The complex is based on the domestic drone BAS ID-100A. It operates at a height of 3-7 m, processing a strip 20-23 m wide. The consumption of agrochemicals is 1-50 liters / ha, depending on the type of preparation and culture. The maximum speed is 70 km/h, the working speed is 25-35 km/h.
Live skin from the laboratory
For the first time, Australian researchers from the Fraser Institute at the University of Queensland have grown a skin organoid that completely mimics human skin. To do this, the researchers used stem cells reprogrammed from skin cells.
Unlike previous models, which were limited to a single layer or cell type, the development includes vessels, immune cells, hair follicles and nerve elements, which makes it as close as possible to natural human skin.
The technology opens up new perspectives in the treatment of rare genetic skin diseases, severe burns and wounds, and can also become a platform for drug testing.
The work has been carried out for more than six years and represents a significant step forward in the field of regenerative medicine and tissue bioengineering.
The first AI-powered computer virus
Researchers from an international company specializing in the development of antivirus programs and computer security solutions (ESET) have reported the discovery of the first known ransomware program using artificial intelligence.
"The PromptLock malware uses OpenAI's gpt-oss:20b model locally via the Ollama API to generate malicious scripts," ESET researchers said on the social network X (ex. Twitter).
Although PromptLock has not yet been documented in actual attacks, the discovery demonstrates how the use of publicly available AI tools can enhance ransomware and other cyber threats.
The researchers also noted that the program is written in the Golang language, which is cross-platform and is used in both Windows and Linux.
Переведено сервисом «Яндекс Переводчик»
