This week’s list of data news highlights covers March 29-April 4 and includes articles on a major data release from the Centers for Medicare and Medicaid Services and a map of gene activity in the fetal human brain.
The Centers for Medicare and Medicaid Services announced this week that it will soon publish a large dataset on Medicare payments to individual doctors, with the goal of rooting out fraud and cutting costs in the federal health payment system. The data will include how many times a provider carried out a particular service, how much they charged Medicare for it, and other information. The data includes around 880,000 individual health care providers, but will not include any individual patient data.
This hurricane season, the National Hurricane Center will offer an interactive map to alert coastal residents of potentially deadly storm surges before they occur. The Potential Storm Surge Flood Map, based on real-time weather forecasts, also takes into account the elevations of affected areas. The map has a resolution of 250 meters, allowing emergency response personnel to determine nearly at the house level how deep water is expected to rise.
The U.S. Food and Drug Administration (FDA) is planning to implement a national system for tracking data on counterfeit and contaminated drugs through the distribution chain. Drug manufacturers will be required to attach a unique product identifier to drug packages, and other entities in the supply chain will be required to verify this identifier for every package that comes through. The FDA, which will implement the complex data collection and sharing effort over the next 10 years, hopes it will speed up the process of recalling products, cutting costs and improving safety in the process.
This week, a team of neuroscientists released the most detailed map ever made of the fetal human brain. The map includes information about the activities of around 20,000 genes, and its creators hope it may help shed light on the biological origins of disorders like autism. The scientists, largely hailing from the Allen Institute for Brain Science in Seattle but aided by colleagues around the country, reported their findings in Nature.
The U.S. Navy is testing a fire-fighting robot that can withstand intense heat and perform a variety of tasks autonomously. The robots will use an augmented vision system to help search for survivors. The devices, called Shipboard Autonomous Fire-Fighting Robots (SAFFiR), are designed to supplement and work with human fire-fighters.
Later this year, NASA will release a database of its more than 1,000 open source software projects. The database will include software that runs climate simulations, rocket guidance systems, and other disparate applications. The effort is part of a government-wide push to open up various agency data sources pursuant to the 2013 executive order on open data.
Internet radio company Pandora has increased the time its customers spend listening by drawing from novel data sources. In addition to collecting ordinary listener feedback, such as instances when a user likes or dislikes a particular song, Pandora can determine when the application is running but the user is no longer listening; the company gathers this data by displaying a simple pop-up asking “are you still listening” when the user has not interacted with the app for a certain period of time. Pandora also chooses not to automate some of the data collection and coding of the songs themselves, with humans trumping machines in determining factors including the mood of a song and the number of vocalists a song features.
In a new survey from enterprise software firm Ipswitch, 85 percent of public sector organizations in the UK said they have no plan in place to manage wearable technology entering the workplace. The company conducted the survey using Freedom of Information requests and found that these organizations were at risk of overloading their networks as more and more employees’ devices enter the office and consume bandwidth. Moreover, 79 percent of government departments are unable to differentiate between wired and wireless devices on their network, meaning that they may not be able to easily identify wearables as the source of the bandwidth consumption if disruption occurs.
A team of historians, computer scientists, and designers have created an interactive visualization of how various commodities moved around the world from 1750-1950. Drawing from over 1.6 million documents, the team mapped terms such as “sugar,” “coffee,” and “cattle,” through time and across geographical locations. The project, a collaboration between universities in Canada and the UK, will be used to aid global trade and economics research.
In February, a team of Italian researchers presented work on a new multi-purpose algorithm to automatically extract information from large, complex data sets. The algorithm, known as KODAMA (which stands for “knowledge discovery by accuracy maximization”) was tested on a large gene expression database, as well as a database of U.S. presidents’ State of the Union speeches, where it was able to classify a speech as pre- or post-Reagan.