System Bits: Oct. 31

Software enables cars to auto-report diagnostics
Thanks to researchers at MIT, it may soon be possible to hop into a ride-share car, glance at a smartphone app, and tell the driver that the car’s left front tire needs air, its air filter should be replaced next week, and its engine needs two new spark plugs.

A new smartphone app analyzes a car’s sounds and vibrations, as measured by the phone’s microphone and accelerometers. “We’re listening to the car’s breathing, and listening for when it starts to snore,” MIT research scientist Joshua Siegel says. 
Source: MIT

This capability may be available within the next year or two, in any car someone happens to be in, based on analysis of the car’s sounds and vibrations, as measured by the phone’s microphone and accelerometers.

The MIT research behind this idea has been reported in a series of papers, most recently in the November issue of the journal Engineering Applications of Artificial Intelligence.

The team estimates a smartphone app combining the various diagnostic systems they developed could save the average driver $125 a year and improve their overall gas mileage by a few percentage points. For trucks, the savings could run to $600 a year, not counting the benefits of avoiding breakdowns that could result in lost income.

This is possible because the sensitivity of today’s smartphones is so high, they can do a good job of detecting the relevant signals without needing any special connection. For some diagnostics, though, mounting the phone to a dashboard holder would improve the level of accuracy. Already, the accuracy of the results from the diagnostic systems they have developed are reportedly well in excess of 90 percent. Tests for misfire detection have produced no false positives where a problem was incorrectly identified.

The basic idea is to provide diagnostic information that can warn the driver of upcoming issues or needed routine maintenance, before these conditions lead to breakdowns or blowouts, the researchers added.

Many of the diagnostics are derived by using machine-learning processes to compare many recordings of sound and vibration from well-tuned cars with similar ones that have a specific problem. The machine learning systems can then extract even very subtle differences. For example, algorithms designed to detect wheel balance problems did a better job at detecting imbalances than expert drivers from a major car company, Siegel says.

A prototype smartphone app that incorporates all these diagnostic tools is being developed and should be ready for field testing in about six months, with a commercial version available about a year after that. The system will be commercialized by a startup company Siegel founded called Data Driven.

Fighting cancer with open source machine learning
Georgia Tech researchers who’ve built a program that predicts cancer drug effectiveness via machine learning and raw genetic data would like cancer fighters to take it for free, or even just swipe parts of their programming code — so they’ve made it open source.

The team hopes to attract a crowd of researchers who will also share their own cancer and computer expertise and data to improve upon the program and save more lives together.

A dying cancer cell with filopodia stretched out to its right. The protrusions help cancer migrate. Stock NIH NCMIR image. The image does not display a cell treated in the Georgia Tech study.
Source: NIH-funded image of HeLa cell / National Center for Microscopy and Imaging Research

The researchers said their invitation to take the code is also a gauntlet: they’re challenging others to come beat them at their own game and help hone a formidable software tool for the greater good. Not only the labor but also the fruits will remain openly accessible to benefit the treatment of patients as best possible.

Researchers wanting to participate can follow this link to a new study, and from there they will find links to download the software from GitHub and to access the code.

The program uses proven machine learning mechanisms and also normalizes data. The latter allows the machine learning to work with data from varying sources by making them compatible.

In addition, one big bias the researchers tossed out was a concentration only on gene expression data pertaining to the specific type of cancer they were aiming to treat since they found out it’s better to give the program data from a broad diversity of cancers, and that will actually later give a better prediction of drug effectiveness for a specific cancer like breast cancer.

The researchers also want the project to pool large amounts of anonymous patient treatment success and failure data, which will help the program optimize predictions for everyone’s benefit. But that doesn’t mean some companies can’t benefit, too.

Making fiber optics obsolete
A team of researchers from the University of Glasgow, the Max Planck Institute for the Science of Light and Institute of Optics, and the Universities of Otago, Ottawa and Rochester has taken an important step towards using ‘twisted’ light as a form of wireless, high-capacity data transmission which could make fiber-optics obsolete with ‘optical angular momentum’ (OAM) that they say could overcome current difficulties with using twisted light across open spaces.

The researchers reminded that scientists can ‘twist’ photons – individual particles of light – by passing them through a special type of hologram, similar to that on a credit card, giving the photons a twist known as optical angular momentum. While conventional digital communications use photons as ones and zeroes to carry information, the number of intertwined twists in the photons allows them to carry additional data – something akin to adding letters alongside the ones and zeroes. The ability of twisted photons to carry additional information means that optical angular momentum has the potential to create much higher-bandwidth communications technology.

Source: University of Glasgow

While optical angular momentum techniques have already been used to transmit data across cables, transmitting twisted light across open spaces has been significantly more challenging for scientists to date. Even simple changes in atmospheric pressures across open spaces can scatter light beams and cause the spin information to be lost.

The researchers examined the effects on both the phase and intensity of OAM carrying light over a real link in an urban environment to assess the viability of these modes of quantum information transfer.

Their free space link, in Erlangen, Germany, was 1.6km in length and passed over fields and streets and close to high-rise buildings to accurately simulate an urban environment and atmospheric turbulence that can disrupt information transfer in space – a thorough approach that will be instrumental in moving OAM research forward.

Conducting this field tests in a real urban environment, has revealed exciting new challenges that will that must be overcome before systems can be made commercially available.  Previous studies had indicated the potential feasibly of OAM communication systems, but had not fully characterized the effects of turbulent air on the phase of the structured light propagating over links of this length.

The researchers believe their findings allow for challenges to be addressed that were not previously observed in developing adaptive optics for quantum information transfer to move closer towards a new age of free space optics that will eventually replace fibre optics as a functional mode of communication in urban environments and remote sensing systems.