Scientists will be able to understand the nature of most diseases within three to five years using artificial intelligence and find the key to their treatment, according to Stéphane Bancel, CEO of the American biotech company Moderna.
Moderna’s CEO believes that AI will enable understanding of many diseases within five years. According to Stéphane Bancel, if we understand a disease, we can find a way to cure it.
“If you understand a disease, you can find a way to cure it. In three to five years, scientists will understand most of the diseases we don’t understand now. This will be a first in human history,” he said at an event organized by the American portal Semafor.
Bancel believes the scientists’ research will help more effectively combat previously difficult-to-treat diseases. “The reason people still die from cancer and suffer from Alzheimer’s is because we don’t understand the biology of these diseases,” he noted.
Artificial intelligence tools will help pharmaceutical companies develop new drugs and allow specialists to better diagnose conditions, such as heart disease, Bancel believes. According to the company’s CEO, Moderna scientists are already using AI tools in scientific research, enabling them to achieve previously unattainable results.
Moderna is developing messenger RNA-based pharmacology. The company has been actively developing a COVID-19 vaccine.
A one-time electronic “tattoo” that measures mental workload could be used to improve safety in high-stress jobs, researchers say.
Scientists have invented an electronic temporary facial tattoo designed to measure the wearer’s mental workload, and hope it could help determine when people in high-stakes jobs such as air traffic control need a break.
Over the past year, several high-profile disasters have highlighted the risks associated with overburdened personnel working beyond their capabilities. One notable tragedy occurred in January 2025, when a mid-air collision between a helicopter and a passenger jet killed 55 people. This occurred while the control tower at Reagan National Airport was experiencing an “unusual situation,” and a single air traffic controller was performing a task normally assigned to multiple controllers.
These incidents highlighted a number of pressing issues, including the need for better systems for assessing mental workload in high-stress workplaces that require intense concentration. The researchers described their new system in a paper published May 29, 2025, in the journal Device.
The electronic tattoo works by recording brain waves through the wearer’s forehead and using this data to assess their mental effort. The creators claim their system provides a faster and more objective approach than existing methods.
Currently, self-reporting is the generally accepted approach to assessing mental workload. In other words, people track their fatigue and report it when they feel exhausted, says study co-author Nanshu Lu, a professor of engineering at the University of Texas (UT) at Austin. Unfortunately, however, people aren’t very good at assessing and evaluating their own mental performance.
For this reason, researchers sought to identify objective physiological markers of mental fatigue by recording brain activity. The least invasive method for monitoring brain activity is electroencephalography (EEG), a method that uses carefully placed electrodes to measure electrical signals pulsating across the scalp.
Scientists claim the new “electronic tattoo” is designed to measure the mental workload of people in high-stress jobs. However, others question whether workers and employers will be receptive to this technology. Device/Huh et al.
Conventional EEG devices look like shower caps, but to ensure a stable connection, a conductive substance, such as a special gel, is usually applied to the scalp. Furthermore, numerous wires run from the cap’s electrodes to the device collecting electrical data, making participants look like they’re wearing electric spaghetti.
Lu’s new system, developed with co-author and colleague Louis Sentis, a professor in the University of Texas School of Engineering, uses a disposable polymer adhesive patch—a “tattoo”—that is individually molded to the user’s face. It is combined with a lightweight battery and an electrode system that records brain waves from the user’s forehead.
Brainwaves have different frequencies, ranging from slower (delta and theta) to faster (alpha, beta, and gamma). Previous studies have linked the degree of mental workload to different patterns of these brainwaves.
Lu and Sentis conducted a small pilot study to test their device. Six participants donned a cap and completed a memory test, the difficulty of which increased with each round. As the difficulty increased, the participants’ delta and theta wave power increased, while alpha, beta, and gamma waves decreased, indicating increased mental workload.
This data was then fed into a machine learning model, which estimated the mental workload experienced by each participant while completing tasks of varying difficulty. The estimates obtained by the model correlated closely with the participants’ self-reported workload, measured using the NASA Task Load Index, a questionnaire-based tool.
The correlation between the electronic tattoo and the NASA index was similar to using a sensor placed over the entire scalp. Users’ sweat actually aided the system, reducing electrical resistance and increasing recording accuracy over time. “The longer you wear it, the better,” said Geert Kouwenberghs, a professor of bioengineering at the University of California, San Diego, who was not involved in the study.
However, the new approach has its drawbacks. Currently, the system can only collect signals from hairless areas of skin, such as the forehead, whereas conventional EEG caps record signals from multiple electrodes located across the entire scalp. The team hopes to overcome this by using an approach that involves printing ink-based patterns directly onto the scalp.
The current system records data in real time, like a standard EEG, but the analysis of this electrical data is performed separately on a computer. Lu noted that workers in high-stress environments would likely benefit from real-time assessment and feedback. The team’s next goal is to embed a neural network—another type of artificial intelligence model—into the tattoo’s Bluetooth chip. This will allow both recording and analysis of data directly within the tattoo itself.
“Whether it’s pilots, drivers, or robotic systems specialists—we’re likely to see an increase in the mental workload of people performing critical tasks,” Sentis said. “I see these sensors being used in real time very soon,” he added.
Kauenbergs agreed that the technology could be ready for commercial use, but he questioned whether employers and employees would be willing to accept such intrusive technology in the name of security.
See also... New method for observing individual atoms confirms century-old quantum mechanical theory |
Researchers have developed a machine learning algorithm that accurately recognizes heart murmurs in dogs – a key sign of heart disease that is particularly common in small breeds such as King Charles Spaniels.
Cambridge University researchers adapted an algorithm originally developed for use in human diagnostics. They found that it could automatically detect and evaluate heart murmurs in dogs using audio recordings from digital stethoscopes. During testing, the algorithm detected heart murmurs with 90% sensitivity, achieving an accuracy similar to that used by experienced cardiologists.
Heart murmurs are the primary symptom of mitral valve disease, the most common heart condition in adult dogs. Approximately one in 30 dogs examined by a veterinarian is found to have a heart murmur, with the rate higher in small breeds and older dogs.
Since mitral valve disease and other heart conditions are very common in dogs, early detection is crucial, as prompt treatment can prolong their lives. The technology developed by the Cambridge team could become an accessible and effective screening tool for primary care veterinarians and improve the quality of life for dogs. The study results were published in the Journal of Veterinary Internal Medicine.
The researchers started with a database of heart murmurs collected from approximately 1,000 human patients and developed a machine learning algorithm to simulate whether a cardiologist detected a heart murmur. They then adapted the algorithm so that it could be used to analyze heart murmurs in dogs. The researchers collected data from nearly 800 dogs undergoing routine cardiac examinations at four specialist veterinary centres in the UK. All dogs underwent a full physical examination and cardiac scan (echocardiogram) by a cardiologist to assess the presence of heart murmurs and detect heart disease, and the heart sounds were recorded using an electronic stethoscope. As a result, the researchers obtained the largest dataset of canine heart sounds ever created.
The researchers refined the algorithm so that it could detect and classify heart murmurs based on audio recordings, and distinguish between murmurs associated with mild disease and those that reflect progressive heart disease requiring further treatment.
An analysis of the algorithm’s performance showed that it agreed with the cardiologist’s assessment more than half the time, and was within one point of the cardiologist’s assessment in 90% of cases. The researchers say this is a promising result, as there is often significant variability in how different veterinarians assess heart murmurs.
The severity of a heart murmur is a useful criterion for determining further steps and treatment methods, and scientists have automated this process. The resulting algorithm could become a valuable tool for veterinarians and nurses. It will help determine when and at what dosage to administer medications, which is crucial for ensuring dogs have the best quality of life for as long as possible.
See also... Complete list of HTML/Javascript event handlers with examples List of required documents for a website in the Russian Federation |
