Skip to content

Category: heart

Light-sensitive bacteria could save you during a heart attack

Heart attacks are frightening by themselves, but they're made worse by the potential for lasting damage. Even a brief interruption to blood flow could permanently destroy vital tissue that keeps your heart beating as usual. However, there might be a way to mitigate or even prevent that damage. Scientists have discovered that a light-sensitive bacteria, synechococcus elongatus, can keep oxygen coming in the midst of a heart attack. Much like a plant, the bacteria both draws on photosynthesis for energy and turns both CO2 and water into oxygen. If you expose it to light soon after the attack, you can maintain oxygen levels and increase the heart's blood-pumping ability after the attack is over.

In lab rats, the results were dramatic. Oxygen levels were 25 times higher 10 minutes after the attack, and the hearts pumped 60 percent more blood 45 minutes after the attack. If you could use this as an emergency treatment in humans, it could mean the difference between outright heart failure and a reasonably healthy patient.

The emphasis is on "if," however. It's easy to shine light into the small body of a rat; it's tougher to do that with humans, who have thicker heart muscles (and are much larger, of course). There's also the question of whether or not the bacteria are completely safe. Don't count on this solution reaching hospitals soon, if at all. Nonetheless, the discovery is promising: it suggests that there's a way to protect your heart against long-term harm even as doctors race to save you from the immediate threat.

Source: Science, Science Advances

Comments closed

AI can predict if you’ll die soon by examining your organs

When not stealing blue collar jobs, robots are becoming Dr. House, diagnosing maladies like breast and skin cancer with aplomb. Scientists at the University of Adelaide have pushed it to a morbid new level, however, with an AI system that can tell if you're going to die. By analyzing CT scans from 48 patients, the deep learning algorithms could predict whether they'd die within five years with 69 percent accuracy -- "broadly similar" to scores from human diagnosticians, the paper says.

Luckily, foretelling such dire consequences may help doctors to stave them off. "Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual," lead author Dr. Luke Oakden-Rayner told the University of Adelaide. "Instead of focusing on diagnosing diseases, the automated systems can predict medical outcomes in a way that doctors are not trained to do, by incorporating large volumes of data and detecting subtle patterns."

For this study, the system was looking for things like emphysema, an enlarged heart and vascular conditions like blood clotting.The deep learning system was trained to analyze over 16,000 image features that could indicate signs of disease in those organs. Machines have become adept at it surprisingly quickly, even though it's "something that requires extensive training for human experts," said Oakden-Rayner.

AI can pick problems in the heart and lungs (at left) that might lead to an early death.

The goal was not to build a grim diagnostic system, and the AI only analyzed retrospective patient data. Rather, the team is looking to lay the groundwork for algorithms that can diagnose your overall health, rather than just spotting a single disease. They also want to "motivate the use of routinely collected, high resolution radiologic images as sources of high quality data for precision medicine," according to the paper. In other words, they're encouraging more scans as a way to improve the results of future diagnostic systems.

"Our research opens new avenues for the application of artificial intelligence technology in medical image analysis, and could offer new hope for the early detection of serious illness, requiring specific medical interventions," says Oakden-Rayner.

Via: Reddit

Source: Nature

Comments closed

AI can predict if you’ll die soon by examining your organs

When not stealing blue collar jobs, robots are becoming Dr. House, diagnosing maladies like breast and skin cancer with aplomb. Scientists at the University of Adelaide have pushed it to a morbid new level, however, with an AI system that can tell if you're going to die. By analyzing CT scans from 48 patients, the deep learning algorithms could predict whether they'd die within five years with 69 percent accuracy -- "broadly similar" to scores from human diagnosticians, the paper says.

Luckily, foretelling such dire consequences may help doctors to stave them off. "Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual," lead author Dr. Luke Oakden-Rayner told the University of Adelaide. "Instead of focusing on diagnosing diseases, the automated systems can predict medical outcomes in a way that doctors are not trained to do, by incorporating large volumes of data and detecting subtle patterns."

For this study, the system was looking for things like emphysema, an enlarged heart and vascular conditions like blood clotting.The deep learning system was trained to analyze over 16,000 image features that could indicate signs of disease in those organs. Machines have become adept at it surprisingly quickly, even though it's "something that requires extensive training for human experts," said Oakden-Rayner.

AI can pick problems in the heart and lungs (at left) that might lead to an early death.

The goal was not to build a grim diagnostic system, and the AI only analyzed retrospective patient data. Rather, the team is looking to lay the groundwork for algorithms that can diagnose your overall health, rather than just spotting a single disease. They also want to "motivate the use of routinely collected, high resolution radiologic images as sources of high quality data for precision medicine," according to the paper. In other words, they're encouraging more scans as a way to improve the results of future diagnostic systems.

"Our research opens new avenues for the application of artificial intelligence technology in medical image analysis, and could offer new hope for the early detection of serious illness, requiring specific medical interventions," says Oakden-Rayner.

Via: Reddit

Source: Nature

Comments closed

A 3D-printed patch could help you recover from a heart attack

Scientists have dreamed of easily patching up heart tissue in the wake of heart attacks, but there are always gotchas: for example, it's no mean feat to replicate the complex structures of real tissue. However, there may be a solution in sight. Researchers have produced a 3D-printed cell patch that can heal scarred heart tissue. The team used laser-based bioprinting to fit stem cells (based on adult human heart cells) to a matrix developed around a 3D scan of heart tissue's native proteins. When those cells grew, the matrix not only replicated the structures of regular heart tissue (down to 1 micron) but started beating in sync. And the early results are very promising.

After the team tested its patch on a mouse, the rodent's heart saw a "significant increase" in functional capacity in the space of 4 weeks. Moreover, it eventually absorbed into the heart -- the team didn't have to perform follow-up operations to make sure it was a good fit.

Naturally, a mouse heart is easier to fix than a much larger human heart. The researchers see this as just a matter of time, though. They believe that human-scale patches should be viable "within the next several years." If so, recovering from a heart attack may just be a matter of implanting some custom-printed tissue and waiting for your health to improve.

Source: University of Minnesota, PubMed.gov

Comments closed

Scientists turn spinach into a farm for human heart cells

In the future, you might not have to resort to exotic materials to create heart cells -- you could just raid your grocery store's produce section. Scientists have invented a process that turns spinach leaves into farms for functioning human heart cells. The team started by pumping a detergent solution through the spinach, stripping it of its plant cells and turning it into a ghostly shell made mostly of cellulose. After that, they cultured heart cells on the remaining structure, sending both fluids and microscopic beads through the vegetable's now-empty veins in order to feed the new cells.

The result is a safe and potentially ideal bed for growing heart tissue. Existing approaches to generating heart cells, such as 3D printing, aren't good at replicating the extremely complex network of veins needed to grow cells -- spinach already has that. And when cellulose is bio-compatible, you don't have to worry about hostile interactions between the heart cells and their host.

As is often the case with this kind of research, it'll be a long while before this method is ready for practical use. If it pans out, though, it could lead to an affordable and eco-friendly of growing replacement tissue for heart attack patients. You could potentially apply this to other body replacements, too -- the researchers suggest that wood's structure might be helpful for generating bone. They've already applied this method to parsley, peanut hairy roots and sweet wormwood, so it's more a question of finding use cases than overcoming technical hurdles.

Via: Ars Technica

Source: WPI, ScienceDirect

Comments closed
%d bloggers like this: