The brain performs very complex computational tasks when identifying what is being seen by the eyes and modern computer vision systems come nowhere near that ability. Scientists from Harvard and MIT are working on adapting computer graphics processors to this specific task in an attempt to reverse engineer some of the characteristics of biologic visual systems.

To tackle this problem, the team drew inspiration from screening techniques in molecular biology, where a multitude of candidate organisms or compounds are screened in parallel to find those that have a particular property of interest. Rather than building a single model and seeing how well it could recognize visual objects, the team constructed thousands of candidate models, and screened for those that performed best on an object recognition task.
The resulting models outperformed a crop of state-of-the-art computer vision systems across a range of test sets, more accurately identifying a range of objects on random natural backgrounds with variation in position, scale, and rotation.
Using ordinary computer processing units, the effort would have required either years of time or millions of dollars of computing hardware. Instead, by harnessing modern graphics hardware, the analysis was done in just one week, and at a small fraction of the cost.
“GPUs are a real game-changer for scientific computing. We made a powerful parallel computing system from cheap, readily available off-the-shelf components, delivering over hundred-fold speed-ups relative to conventional methods,” says Pinto [Nicolas Pinto, Ph.D. candidate at MIT]. “With this expanded computational power, we can discover new vision models that traditional methods miss.”

Duke University scientists demonstrated the possibility of using RNA aptamers to target tumor cells without affecting healthy tissue or activating the immune system. They did this on a mouse model by meticulously searching for a short strand aptemer that would bind to the specific tumor protein:

The researchers used a large pool of RNA strands and applied them to a rodent with a liver tumor, the type of metastatic tumor that often results from a colon cancer tumor.
“We hypothesized that the RNA molecules that bind to normal cellular elements would be filtered out, and this happened,” said Clary [Bryan Clary, MD, chief of the Division of Hepatopancreatobiliary and Oncologic Surgery --ed.], who treats colon cancer patients. “In this way, we found the RNA molecules that went specifically to the tumor.”
The researchers removed the tumor, extracted the specific RNA in the tumor, amplified these pieces of RNA to create a greater amount, and reinjected the molecules to learn which bound most tightly to the tumor. They repeated this process 14 times to find a good candidate.
The team found a tumor-targeting RNA aptamer that specifically bound to RNA helicase p68, a nuclear protein produced in colorectal tumors.
“This aptamer not only binds to p68 protein in cell culture, but also preferentially binds to cancer deposits in a living animal,” Mi said. “The nice thing about this aptamer approach is that it could be used to discover the molecular signatures of many other diseases.”
Clary said the process could be repeated with different types of tumors. For example, a scientist might take a breast cancer line and grow it in the lung as a metastasis model and then perform in vivo selection to identify RNAs specifically binding to the lung tumor.

Full story: First Live Targeting of Tumors with RNA-Based Technology…
Abstract in Nature Chemical Biology: In vivo selection of tumor-targeting RNA motifs
Image credit: Wellcome Images…

Oil excavation firms and cardiovascular researchers spend a lot of time studying how liquids move through vessels. In order to have an exchange of ideas between such different industries, the third Pumps and Pipes symposium will be held next Monday at the University of Houston.

Pumps & Pipes III: Better Together will have speakers in the morning sessions from medicine, energy, and academia discussing use of advanced nanotechnology, robotics and distant monitoring in common issues like pipeline corrosion and blood vessel integrity. The afternoon sessions will feature new discussions on pipes and fluids, a concept that spawned joint oil and medicine ideas in the past when Methodist researchers looking at preventing aneurysms gained a new perspective of blood flow dynamics from pipeline engineers who used fluid dynamics to predict pipeline ruptures. Talks will focus on managing imperfect pipes, next-generation intelligent conduits, and advanced materials for energy and medicine. The presentations are designed to offer common language and terminology to all parties, as well as provide a platform to discuss the hurdles facing each discipline.

Full story from the University of Houston: Similarities of Pumping Blood and Oil Examined at “Pumps and Pipes III”
Link: Pumps and Pipes…