Scientists in Pennsylvania are planning to re-engineer human immune cells to sniff out and kill cancer, a drone saved swimmers caught in rough surf in Australia, and an AI robot assistant in England found a new way to fight malaria — in toothpaste. Time to brush up on these and other remarkable developments we discovered this week.
What is it? Scientists at the University of Pennsylvania are planning to use the gene-editing tool CRISPR to alter the human body’s immune cells so they can recognize and kill cancer. They are planning a clinical study that may involve as many as 18 patients suffering from different cancers, including multiple myeloma, melanoma and synovial sarcoma.
Why does it matter? Researchers already have used viruses to train the immune system to attack cancer, but this would be the first human trial in the U.S. that uses CRISPR to achieve a similar feat.
How does it work? CRISPR allows scientists to rewrite faulty or unwanted human, animal and plant DNA. The team plans to remove the patients’ immune systems T-cells, edit them so they could fight cancer more effectively, and infuse them back into the body.
What is it? An autonomous “lifeguard” drone called Lifesaver rescued two teenage boys in Australia when beachgoers spotted them struggling in rough surf some 2,300 feet from the shore, according to the Sydney Morning Herald. The beach patrol navigated the drone to the pair and dropped an “inflatable rescue pod” they could hold on to.
Why does it matter? Human lifeguards were still learning to use the drone when they deployed it on its maiden rescue mission. The drones, made by Australia’s Westpac Little Ripper company, are part of the local government’s multimillion-dollar “shark mitigation strategy.”
How does it work? The drone is equipped with a camera, digital controls and other technology. Lifeguard supervisor Jai Sheridan told the Herald that the drone enabled him to deliver aid within minutes of receiving the initial alert. “I was able to launch it, fly it to the location, and drop the pod all in about one to two minutes,” he said. “On a normal day that would have taken our lifeguards a few minutes longer to reach the members of the public.” You can see the rescue video here.
What is it? The Chinese city Xian, perhaps best known for being home to the world’s largest army of terracotta soldiers, is now building what might be the world’s biggest air purifier. Engineers already have started testing the concrete and steel structure, which is 323 feet tall.
Why does it matter? Many Chinese cities have been suffering from crippling air pollution, especially in the winter. Cao Junji, who runs the research, said that “improvements in air quality had been observed over an area of 10 square kilometers (3.86 square miles) in the city over the past few months and the tower has managed to produce more than 10 million cubic meters (353 million cubic feet) of clean air a day since its launch,” according to South China Morning Post. “Cao added that on severely polluted days the tower was able to reduce smog close to moderate levels,” the paper wrote.
How does it work? Cao and his team built the tower on the outskirts of town. The structure relies on “greenhouses covering about half the size of a soccer field around the base of the tower,” according to the newspaper. “Polluted air is sucked into the glasshouses and heated up by solar energy. The hot air then rises through the tower and passes through multiple layers of cleaning filters.” Cao told the paper that the tower “barely requires any power input throughout daylight hours. The idea has worked very well in the test run.”
What is it? A “robot scientist” with AI smarts called Eve helped researchers at Cambridge University figure out that triclosan, an ingredient normally used in toothpaste to fight plaque, could be used to fight drug-resistant malaria. The research was published in the journal Scientific Reports.
Why does it matter? Although doctors are coming up with new ways to treat malaria, the disease still kills more than 500,000 people each year, according to the researchers. Scientists are growing worried that strains of the malaria parasite may grow immune to these drugs. “The discovery by our robot ‘colleague’ Eve that triclosan is effective against malaria targets offers hope that we may be able to use it to develop a new drug,” said Elizabeth Bilsland, the lead author of the paper, who is now an assistant professor at the University of Campinas, Brazil. “We know it is a safe compound, and its ability to target two points in the malaria parasite’s lifecycle means the parasite will find it difficult to evolve resistance.”
How does it work? Eve helped the team figure out that triclosan “affects parasite growth by specifically inhibiting an entirely different enzyme of the malaria parasite, called DHFR,” the University of Cambridge said in a news release. “DHFR is the target of a well-established antimalarial drug, pyrimethamine; however, resistance to the drug among malaria parasites is common, particularly in Africa. The Cambridge team showed that triclosan was able to target and act on this enzyme even in pyrimethamine-resistant parasites.”
What is it? More AI news popped up on this side of the ocean. Researchers at Stanford University used radiological images from more than 12,000 patients to train a neural network to “detect and localize abnormalities.” The images of wrists, elbows, shoulders and other body parts came from studies labeled by human doctors as normal or abnormal. The team than matched the neural network against a team of certified Stanford radiologists to test its performance. How did it do? “We find that our model achieves performance comparable to that of radiologists,” the team reported in the online journal arXiv. The team wrote that that model achieved “higher than the best radiologist performance in detecting abnormalities on finger studies and equivalent on wrist studies. However, model performance is lower than best radiologist performance in detecting abnormalities on elbow, forearm, hand, humerus, and shoulder studies, indicating that the task is a good challenge for future research,” the team said.
Why does it matter? The team wrote that “there has been a growing effort to make repositories of medical radiographs openly available,” and that “access to large data sets “have led to deep learning algorithms achieving or approaching human-level performance on tasks such as image recognition.” Similar algorithms could be help doctors prioritize their work flow, moving the most urgent cases to the top of the pile. Software could also “help combat radiologist fatigue,” the team wrote.
How does it work? The team wrote that they built “a 169-layer convolutional neural network to predict the probability of abnormality for each image in a study.” The network’s architecture “connects each layer to every other layer in a feed-forward fashion to make the optimization of deep networks tractable.” Late last year, another team at Stanford achieved a similar feat when they software to analyze chest X-rays and detect pneumonia.