Novel signature could improve understanding of depression and its subtypes, and how to treat them

Using machine learning, researchers have identified novel, distinct patterns of coordinated activity between different parts of the brain in people with major depressive disorder--even when different protocols are used to detect these brain networks. Ayumu Yamashita of Advanced Telecommunications Research Institutes International in Kyoto, Japan, and colleagues present these findings in the open-access journal PLOS Biology. 

While major depression is usually straightforward to diagnose, a better understanding of the brain networks associated with depression could improve treatment strategies. Machine-learning algorithms can be applied to data on brain activity in people with depression in order to find such associations. However, most studies have focused only on specific subtypes of depression, or they have not accounted for the differences in brain imaging protocols between healthcare institutions. 

To address these challenges, Yamashita and colleagues used machine learning to analyze brain network data from 713 people, 149 of whom had major depression. These data had been collected using a technique called resting-state functional MRI (rs-fMRI), which detects brain activity and produces images that reveal coordinated activity, or "functional connections," between different parts of the brain. The imaging had been performed at different institutions using different protocols. {module INSIDE STORY}

 The machine-learning method identified key functional connections in the imaging data that could serve as a brain network signature for major depression. Indeed, when the researchers applied that new signature to rs-fMRI data collected at different institutions from 521 other people, they achieved 70 percent accuracy in identifying which of those new people had major depressive disorder.

The researchers hope that their new brain network signature, which can be applied across different imaging protocols, could serve as a foundation for discovering brain network patterns associated with subtypes of depression, and to reveal relationships between depression and other disorders. A better understanding of brain network connections in major depression could help match patients to effective treatments and inform development of new treatments.

A new study of airflow patterns inside a car's passenger cabin offers some suggestions for potentially reducing the risk of COVID-19 transmission while sharing rides with others.

The study, by a team of Brown University researchers, used supercomputer models to simulate the airflow inside a compact car with various combinations of windows open or closed. The simulations showed that opening windows -- the more windows the better -- created airflow patterns that dramatically reduced the concentration of airborne particles exchanged between a driver and a single passenger. Blasting the car's ventilation system didn't circulate air nearly as well as a few open windows, the researchers found.

"Driving around with the windows up and the air conditioning or heat on is definitely the worst scenario, according to our computer simulations," said Asimanshu Das, a graduate student in Brown's School of Engineering and co-lead author of the research. "The best scenario we found was having all four windows open, but even having one or two open was far better than having them all closed."  {module INSIDE STORY}

Das co-led the research with Varghese Mathai, a former postdoctoral researcher at Brown who is now an assistant professor of physics at the University of Massachusetts, Amherst. The study is published in the journal of Science Advances.

The researchers stress that there's no way to eliminate risk completely -- and, of course, current guidance from the U.S. Centers for Disease Control (CDC) notes that postponing travel and staying home is the best way to protect personal and community health. The goal of the study was simply to study how changes in airflow inside a car may worsen or reduce the risk of pathogen transmission.

The computer models used in the study simulated a car, loosely based on a Toyota Prius, with two people inside -- a driver and a passenger sitting in the back seat on the opposite side from the driver. The researchers chose that seating arrangement because it maximizes the physical distance between the two people (though still less than the 6 feet recommended by the CDC). The models' simulated airflow around and inside a car moving at 50 miles per hour, as well as the movement and concentration of aerosols coming from both driver and passenger. Aerosols are tiny particles that can linger in the air for extended periods of time. They are thought to be one way in which the SARS-CoV-2 virus is transmitted, particularly in enclosed spaces.

Part of the reason that opening windows is better in terms of aerosol transmission is that it increases the number of air changes per hour (ACH) inside the car, which helps to reduce the overall concentration of aerosols. But ACH was only part of the story, the researchers say. The study showed that different combinations of open windows created different air currents inside the car that could either increase or decrease exposure to remaining aerosols.

Because of the way air flows across the outside of the car, air pressure near the rear windows tends to be higher than the pressure at the front windows. As a result, air tends to enter the car through the back windows and exit through the front windows. With all the windows open, this tendency creates two more-or-less independent flows on either side of the cabin. Since the occupants in the simulations were sitting on opposite sides of the cabin, very few particles end up being transferred between the two. The driver in this scenario is a slightly higher risk than the passenger because the average airflow in the car goes from back to front, but both occupants experience a dramatically lower transfer of particles compared to any other scenario.

The simulations for scenarios in which some but not all windows are down yielded some possibly counterintuitive results. For example, one might expect that opening windows directly beside each occupant might be the simplest way to reduce exposure. The simulations found that while this configuration is better than no windows down at all, it carries a higher exposure risk compared to putting down the window opposite each occupant.  {module INSIDE STORY}

"When the windows opposite the occupants are open, you get a flow that enters the car behind the driver, sweeps across the cabin behind the passenger and then goes out the passenger-side front window," said Kenny Breuer, a professor of engineering at Brown and a senior author of the research. "That pattern helps to reduce cross-contamination between the driver and passenger."

It's important to note, the researchers say, those airflow adjustments are no substitute for mask-wearing by both occupants when inside a car. And the findings are limited to potential exposure to lingering aerosols that may contain pathogens. The study did not model larger respiratory droplets or the risk of actually becoming infected by the virus.

Still, the researchers say the study provides valuable new insights into air circulation patterns inside a car's passenger compartment -- something that had received little attention before now.

"This is the first study we're aware of that really looked at the microclimate inside a car," Breuer said. "There had been some studies that looked at how much external pollution gets into a car, or how long cigarette smoke lingers in a car. But this is the first time anyone has looked at airflow patterns in detail."

The research grew out of a COVID-19 research task force established at Brown to gather expertise from across the University to address widely varying aspects of the pandemic. Jeffrey Bailey, an associate professor of pathology and laboratory medicine and a co-author of the airflow study, leads the group. Bailey was impressed with how quickly the research came together, with Mathai suggesting the use of computer simulations that could be done while laboratory research at Brown was paused for the pandemic.

"This is really a great example of how different disciplines can come together quickly and produce valuable findings," Bailey said. "I talked to Kenny briefly about this idea, and within three or four days his team was already doing some preliminary testing. That's one of the great things about being at a place like Brown, where people are eager to collaborate and work across disciplines."