Open source is fertile ground for transformative software, especially in cutting-edge domains like artificial intelligence (AI) and machine learning. The open source ethos and collaboration tools make it easier for teams to share code and data and build on the success of others.
This article looks at 13 open source projects that are remaking the world of AI and machine learning. Some are elaborate software packages that support new algorithms. Others are more subtly transformative. All of them are worth a look.
Neurodegenerative diseases—like amyotrophic lateral sclerosis (ALS, or Lou Gehrig’s disease), Alzheimer’s, and Parkinson’s—are complicated, chronic ailments that can present with a variety of symptoms, worsen at different rates, and have many underlying genetic and environmental causes, some of which are unknown. ALS, in particular, affects voluntary muscle movement and is always fatal, but while most people survive for only a few years after diagnosis, others live with the disease for decades. Manifestations of ALS can also vary significantly; often slower disease development correlates with onset in the limbs and affecting fine motor skills, while the more serious, bulbar ALS impacts swallowing, speaking, breathing, and mobility. Therefore, understanding the progression of diseases like ALS is critical to enrollment in clinical trials, analysis of potential interventions, and discovery of root causes.
However, assessing disease evolution is far from straightforward. Current clinical studies typically assume that health declines on a downward linear trajectory on a symptom rating scale, and use these linear models to evaluate whether drugs are slowing disease progression. However, data indicate that ALS often follows nonlinear trajectories, with periods where symptoms are stable alternating with periods when they are rapidly changing. Since data can be sparse, and health assessments often rely on subjective rating metrics measured at uneven time intervals, comparisons across patient populations are difficult. These heterogenous data and progression, in turn, complicate analyses of invention effectiveness and potentially mask disease origin.
In a new paper published today in the journal Nature Ecology and Evolution, scientists have estimated the conservation status of nearly 1,900 palm species using artificial intelligence, and found more than 1,000 may be at risk of extinction.
The international team of researchers from the Royal Botanic Gardens, Kew, the University of Zurich, and the University of Amsterdam, combined existing data from the International Union for Conservation of Nature (IUCN) Red List with novel machine learning techniques to paint a clearer picture of how palms may be threatened. Although popular and well represented on the Red List, the threat to some 70% of these plants has remained unclear until now.
The IUCN Red List of Threatened Species is widely considered to be a gold standard for evaluating the conservation status of animal, plant, and fungal species. But there are gaps in the Red List that need to be addressed, as not all species have been listed and many of the assessments are in need of an update. Conservation efforts are further complicated by inadequate funding, the sheer amount of time needed to manually assess a species, and public perception favoring certain vertebrate species over plants and fungi.
Daegu Gyeongbuk Institute of Science & Technology (DGIST, President Yang Kook) Professor Hongsoo Choi’s team of the Department of Robotics and Mechatronics Engineering collaborated with Professor Sung-Won Kim’s team at Seoul St. Mary’s Hospital, Catholic University of Korea, and Professor Bradley J. Nelson’s team at ETH Zurich to develop a technology that produces more than 100 microrobots per minute that can be disintegrated in the body.
Microrobots aiming at minimal invasive targeted precision therapy can be manufactured in various ways. Among them, ultra-fine 3D printing technology called two-photon polymerization method, a method that triggers polymerization by intersecting two lasers in synthetic resin, is the most used. This technology can produce a structure with nanometer-level precision. However, a disadvantage exists in that producing one microrobot is time consuming because voxels, the pixels realized by 3D printing, must be cured successively. In addition, the magnetic nanoparticles contained in the robot can block the light path during the two-photon polymerization process. This process result may not be uniform when using magnetic nanoparticles with high concentration.
To overcome the limitations of the existing microrobot manufacturing method, DGIST Professor Hongsoo Choi’s research team developed a method to create microrobots at a high speed of 100 per minute by flowing a mixture of magnetic nanoparticles and gelatin methacrylate, which is biodegradable and can be cured by light, into the microfluidic chip. This is more than 10,000 times faster than using the existing two-photon polymerization method to manufacture microrobots.
The joint research team of Professor Choi Hongsoo at Robotics Engineering, DGIST, a senior researcher Jinyoung Kim from DGIST-ETH Microrobotics Research Center, and the research team of Professor Sung Won Kim at Seoul St. Mary’s Hospital of the Catholic University, made a breakthrough for the improvement of the therapeutic efficacy and safety in stem cell-based treatments.
The team developed a magnetically powered human nuclear transfer stem cells (hNTSC)-based microrobot and a method of minimally invasive delivery of therapeutic agents into the brain via the intranasal pathway. And they also accomplished transplanting the developed stem cell-based microrobot into brain tissue through the intranasal pathway that bypasses the blood-brain barrier. The proposed method is superior in efficacy and safety compared to the conventional surgical method and is expected to bring new possibilities of treating various intractable neurological diseases such as Alzheimer’s disease, Parkinson’s disease, and brain tumors, in the future.
The limitation of stem cell therapy is the difficulty in delivering an exact amount of stem cells to an accurate targeted location deep in the body where the treatment is with high risk. Another limitation is that both efficacy and safety of the treatment are low owing to a large amount of the therapeutic agent loss during delivery, while the cost of the treatment is high. In particular, when delivering stem cells into the brain through blood, the efficiency of cell delivery may decrease owing to the “blood-brain barrier,” which is a unique and specific component of the cerebrovascular network.
One day they shall make nano bots out of graphine.
A team of researchers affiliated with several institutions in South Korea has created microrobots that are able to serve as bridge builders between rat nerve cell networks. In their paper published in the journal Science Advances, the group describes how their microrobots were constructed and how well they served as a bridge builder between neural networks.
Scientists have taken many approaches to study of the brain. One way is to try to grow a brain from nerve cells. Prior work has shown that it is possible to grow a network of neural cells on a glass plate Such a network is, of course, 2-D. In this new effort, the researchers have taken a step toward the creation of a 3D neural network by devising a way to connect 2-D neural networks using microrobots.
Nanoengineers at the University of California San Diego have developed microscopic robots, called microrobots, that can swim around in the lungs, deliver medication and be used to clear up life-threatening cases of bacterial pneumonia.
In mice, the microrobots safely eliminated pneumonia-causing bacteria in the lungs and resulted in 100% survival. By contrast, untreated mice all died within three days after infection.
The results are published Sept. 22 in Nature Materials.
Jülich researchers have been able to demonstrate an exotic electronic state, so-called Fermi Arcs, for the first time in a 2D material. The surprising appearance of Fermi arcs in such a material provides a link between novel quantum materials and their respective potential applications in a new generation of spintronics and quantum computing. The results have recently been published in Nature Communications.
The newly detected Fermi arcs represent special—arc-like—deviations from the so-called Fermi surface. The Fermi surface is used in condensed matter physics to describe the momentum distribution of electrons in a metal. Normally, these Fermi surfaces represent closed surfaces. Exceptions such as the Fermi arcs are very rare and often are associated with exotic properties like superconductivity, negative magnetoresistance and anomalous quantum transport effects.
Today’s technology challenge is to develop the “on-demand” control of physical properties in materials. However, such experimental tests have been largely limited to bulk materials and are key grand challenges in condensed matter science. With its groundbreaking paradigm, the findings present a promising new frontier for quantum control of topological states in low-dimensional systems by external means—the external magnetic field that offers unprecedented capabilities on 2D materials for artificial intelligence as well as future information processing.
PRESS RELEASE — There has been a lot of buzz about quantum computers and for good reason. The futuristic computers are designed to mimic what happens in nature at microscopic scales, which means they have the power to better understand the quantum realm and speed up the discovery of new materials, including pharmaceuticals, environmentally friendly chemicals, and more. However, experts say viable quantum computers are still a decade away or more. What are researchers to do in the meantime?
A new Caltech-led study in the journal Science describes how machine learning tools, run on classical computers, can be used to make predictions about quantum systems and thus help researchers solve some of the trickiest physics and chemistry problems. While this notion has been shown experimentally before, the new report is the first to mathematically prove that the method works.
“Quantum computers are ideal for many types of physics and materials science problems,” says lead author Hsin-Yuan (Robert) Huang, a graduate student working with John Preskill, the Richard P. Feynman Professor of Theoretical Physics and the Allen V. C. Davis and Lenabelle Davis Leadership Chair of the Institute for Quantum Science and Technology (IQIM). “But we aren’t quite there yet and have been surprised to learn that classical machine learning methods can be used in the meantime. Ultimately, this paper is about showing what humans can learn about the physical world.”