It's that time of the week to catch up with all the latest AI and Data Science top stories from the past seven days...
AI detects brain tumours in under three minutes
A recent study has shown that AI can diagnose a brain tumour in a tenth of the time that it takes now, while also detecting some details that conventional practices can miss.
The method works by streamlining the practice of analyzing tissue samples, while the patient is still on the operating table.
While the tech was only marginally more accurate than current methods, which requires sending the tissue to a lab, freezing and staining it, it did slash the time it takes to make the diagnoses from 20-30 minutes to under three minutes.
MIT Develops Machine-Learning Tool to Make Code Run Faster
MIT researchers have built a new benchmark tool that can accurately predict how long it takes given code to execute on a computer chip, which can help programmers tweak the code for better performance.
The tool now makes it easier to quickly learn performance speeds for any new chip architectures. For instance, domain-specific architectures, such as Google’s new Tensor Processing Unit used specifically for neural networks, are now being built but aren’t widely understood.
Google's AI system detects breast cancer more accurately than current methods
A study that tested the accuracy of the system, which was developed through a collaboration between the tech giant and cancer researchers, was published last week in the scientific journal Nature.
The program was trained to detect cancer using tens of thousands of mammograms from women in the United Kingdom and the United States. Early research shows it can produce more accurate detection than human radiologists.
According to the study, using the AI technology resulted in fewer false positives, where test results suggest cancer is present when it isn't, and false negatives, where existing cancer goes undetected.
Startup aims to help stroke survivors talk to smart homes
Voiceitt is integrating their nonstandard speech recognition technology into voice-activated devices. Voiceitt’s smart home system can’t understand everything people with nonstandard speech say. Users train the system with their own voice and with key phrases like “turn the light on,” which lets it learn each person’s specific vocal patterns.
Many people who suffer from medical conditions like Parkinson's, cerebral palsy and stroke would benefit from voice-activated technologies like smart homes, but standard systems often can’t understand their speech. Improving popular smart devices’ accuracy with nonstandard speech is an important goal to help bring technology to people who may benefit the most.
Facebook to ban 'Deepfakes'
Facebook has announced it will remove videos modified by Artificial Intelligence, known as deepfakes, from its platform.
Deep fakes are highly realistic computer-generated clips that are designed to look real and is widely considered dangerous because of the ease in which it can be used to spread misinformation.
Facebook said it would remove videos if it realised they had been edited in ways that weren't obvious to an average person, or if they misled a viewer into thinking that a person in a video said words they did not actually say.
More students interested in AI and Machine Learning now more than ever.
According to a recent report published by Stanford University, more and more students are enrolling in introductory AI and Machine Learning classes. At Stanford itself, enrollment in the school’s “Introduction to Artificial Intelligence” course has grown “fivefold” between 2012 and 2018.