Researchers at the University of Pittsburgh School of Medicine and Carnegie Mellon University College of Engineering have created a machine-learning algorithm that can detect subtle signs of osteoarthritis--too abstract to register in the eye of a trained radiologist--on an MRI scan taken years before symptoms even begin. These results will publish this week in PNAS. With this predictive approach, patients could one day be treated with preventative drugs rather than undergoing joint replacement surgery. "The gold standard for diagnosing arthritis is X-ray. As the cartilage deteriorates, the space between the bones decreases," said study co-author Kenneth Urish, M.D., Ph.D., associate professor of orthopaedic surgery at Pitt and associate medical director of the bone and joint center at UPMC Magee-Womens Hospital.

At least half of the world's population cannot obtain essential health services. But low-cost, easy-to-use technologies powered by Artificial Intelligence (AI) promise to deliver quality and affordable health care to the people who need it most, no matter how hard to reach. At the AI for Good Global Summit last week, entrepreneurs, AI experts, academics and UN representatives described many AI technologies for health, allowing for the early detection of various pathologies such as osteoarthritis, diabetic retinopathy, child malnutrition, snakebites and others. These technologies don't place a heavy burden on doctors, and can lead to prompt diagnosis and effective treatment. They agreed that AI can add tremendous value in developing countries where there is a low density of physicians.

The technology, called Stoplift, analyzes security video to automatically detect theft or errors at the checkout, according to Malay Kundu, the creator of Stoplift. "It can actually tell what you've handled versus what you've rung up," Kundu said. The Cambridge businessman used to develop real-time facial recognition systems to look for terrorists in airports. He realized similar technology could be used at the checkout to tackle a $13 billion per-year problem for grocery stores in the United States. "For every item that is stolen, they have to sell 50 more just to make up for that one item that was lost," Kundu said.

Some grocery stores in Rhode Island and Massachusetts are using artificial intelligence to catch shoplifters. The technology, called Stoplift, analyzes security video to automatically detect theft or errors at the checkout, according to Malay Kundu, the creator of Stoplift. "It can actually tell what you've handled versus what you've rung up," Kundu said. The Cambridge businessman used to develop real-time facial recognition systems to look for terrorists in airports. He realized similar technology could be used at the checkout to tackle a $13 billion per-year problem for grocery stores in the United States.

At a meeting here in Austin, Evernote CTO Anirban Kundu showed Engadget an early version of Spaces, which is meant to facilitate easier collaboration, but also uses AI to deliver better search results and suggest relevant tasks. Evernote said the demo is slow, though, and that the eventual version should perform faster. The note-taking app uses a method that feels similar to what Google Inbox does, by scanning the text in your documents to look for action items like, "Terrence should remind writers to send their drafts by Monday." I typed "Taylor needs to go see a doctor tomorrow" into an early version of the text editor, and within seconds a star icon popped up on the left of my sentence. Despite Evernote's warning about speed, it already seemed pretty quick during my preview.

Evernote decided last year that it wanted to move away from running its own data centers and start using the public cloud to operate its popular note-taking service. On Wednesday, it announced that the lion's share of the work is done, save for some last user attachments. The company signed up to work with Google, and as part of the migration process, the tech titan sent a team of engineers (in one case, bearing doughnuts) over to work with its customer on making sure the process was a success. Evernote wanted to take advantage of the cloud to help with features based on machine learning that it has been developing. It also wanted to leverage the flexibility that comes from not having to run a data center.

Google has been offering public cloud services for several years now, but the company has continued to lag behind Amazon and Microsoft in customer growth. Under the leadership of VMware co-founder Diane Greene, who serves as the executive vice president of Google Cloud Enterprise, the tech titan has focused harder on forging partnerships and developing products to appeal to large customers. It has added a number of key customers under Greene's tenure, including Spotify. One such win is Evernote, which announced Tuesday it would be migrating its service away from its private data centers and to Google's public cloud. When Evernote was looking for a public cloud provider, the company was interested in not only the base level infrastructure available, but also high-level machine learning services and services for building machine learning-driven systems, said Anirban Kundu, Evernote's CTO.

Evernote CEO Chris O'Neill has had a long couple of days. The company he runs recently ignited a firestorm among its users when it announced a privacy policy change that would have required users to open up all their notes for analysis in order to take advantage of forthcoming machine learning features. "We let our users down," he said in an interview. "We really tactically communicated in about as poor a way as we could." Evernote is going back to the drawing board and reversing course on the proposed policy.

Evernote is trying to improve its digital-note-taking service, but it needs to access and read customer notes to do so. In a corporate blog post published this week, the company said it's updating the terms of its privacy policy to include a change that lets its employees access user notes to improve its machine-learning technologies. Machine learning technologies can allow powerful software to sift through data to find patterns and better automate tasks. Evernote said the decision to have some of its employees read user notes, effective Jan. 27, 2017, was done to "make sure that our machine learning technologies are working correctly, in order to surface the most relevant content and features to you." The company elaborated that although its current "computer systems do a pretty good job, sometimes a limited amount of human review is simply unavoidable in order to make sure everything is working exactly as it should."

Today, the company is shifting gears on the question of how it will keep hold of and track that information: Evernote is migrating all of its data, including some 5 billion notes, to Google's Cloud Platform. As part of that, it will also start to use Google's machine learning APIs to help access and use that data in different ways. As a result, Evernote will be shutting down its previous storage architecture that was based around a private cloud infrastructure, along with some of its own tech. Evernote's new CTO Anirban Kundu told me the first two areas that will be replaced by Google's machine learning APIs are its voice recognition for speech-to-text translations; and natural language processing, used to help search for contextual content. Evernote says it will start the migration to Google in early October, with "a complete migration by the end of 2016."