KW 14: AI recognizes virus infection from a cough, Intelligent cameras to fight coronavirus, AI expert on data protection for coronavirus apps


AI recognizes virus infection from a cough: Researchers at the University of Massachusetts Amherst have developed a portable surveillance tool called FluSense that leverages machine learning and real-time data to monitor flu-like illnesses and flu patterns. The device can detect coughing sounds and crowd size in real time and could add to the collection of tools used to forecast seasonal flu and other viral outbreaks. Researchers placed FluSense devices in four healthcare waiting rooms at UMass’s University Health Services clinic, collecting and analyzing more than 350,000 thermal images and 21 million non-speech audio samples. The results showed that the platform was able to accurately predict daily illness rates at the university clinic.

AI helps organize specialist conferences: Researchers from Fraunhofer Austria in Graz are trying to simplify the organization and selection of contributions at scientific conferences with the help of artificial intelligence. Several hundred publications are often submitted at large conferences, which must be forwarded to suitable reviewers. Based on reviewers’ past publications, the AI learns which subject a reviewer is suitable for. But it’s not just specialist conferences that AI could be useful for. The developers of the program are also considering an application in the area of personnel or patent applications.

Intelligent light thanks to AI: The OpenLicht project, a cooperation between Infineon and various universities, has published its first results. The project researchers developed an AI-based lighting system that adapts the light to the user’s position, amongst other things. The system also has the ability to learn a user’s unique preferences. But what is extraordinary about the OpenLicht system is that the application doesn’t require an internet connection. This is part of the developers’ attempt to ensure privacy protection.

Intelligent cameras to fight coronavirus: The Innsbruck start-up Swarm Analytics wants to limit the spread of the coronavirus using AI-controlled cameras. The cameras are installed in public and designed to check how many people are present in a particular location or driving in a car. Swarm Analytics uses this information to generate statistics and identify places where a particularly large number of people are present at any given time. The recorded images are not permanently saved.

Uber releases machine learning library: Mobility service provider Uber has released a machine learning library called Fiber. The aim of Fiber is to enable the implementation of projects that normally require high computing power. An increase in computation underlies many recent advances in machine learning, with more and more algorithms relying on distributed training for processing an enormous amount of data. But reinforcement and population-based methods pose challenges for reliability, efficiency, and flexibility that some frameworks fall short of satisfying. Fiber addresses these challenges with a lightweight strategy to handle task scheduling. It leverages cluster management software for job scheduling and tracking, doesn’t require preallocating resources, and can dynamically scale up and down on the fly, allowing users to migrate from one machine to multiple machines seamlessly.

Nvidia DLSS 2.0: AI upscaling to significantly increase frame rate in games
Medicine: Artificial intelligence against coronavirus
ATEcare: Artificially intelligent robot inspects assemblies
Health: How two start-ups are fighting coronavirus using data
EU Commission: Intelligent competitive advantages


Artificial intelligence company is teaming up with top universities and companies and spending $367 million in the consortium’s initial five years, aiming its first awards at finding ways to slow the coronavirus pandemic.


AI expert on data protection for coronavirus apps: AI expert Björn Schuller spoke in an interview about the use of apps to contain the coronavirus. Schuller himself is working on an application that uses coughing noises to recognize whether a person has a cold or not. He describes donating data as almost as important as donating blood. While he admits that the use of AI in healthcare is an “ethical gray area”, the data is still needed to develop better artificial intelligence. With the increase in use of health apps, it is important that the data collected is evaluated on the device used. While the use of apps against the coronavirus is still being discussed in Germany, the “stop corona app” is already available in Austria. This app, however, sends the health information it collects to a server in order to process it there.

AI in cybersecurity: Artificial intelligence is increasingly being used in cybersecurity. In the so-called Security Operations Center (SOC), AI can be used to detect data anomalies in real time and thus prevent cyber attacks. Botnets can be identified early by analyzing past communication flows. It is crucial to use the right training material when using AI in IT security. The better the data sets used to train an AI, the earlier and better it can react to possible threats. That is why providers of SOCs are investing heavily in the purchase of global data sets.


Artificial intelligence detects bomb craters: Researchers from Ohio State University (OSU) have used artificial intelligence to detect Vietnam War-era bomb craters in Cambodia from satellite images – with the hope that it can help find unexploded bombs. The new method increased true bomb crater detection by more than 160 percent over standard methods. The researchers used a type of artificial intelligence called machine learning to analyze the satellite images for evidence of bomb craters. After the machine learned how to detect true bomb craters, one of the researchers checked the computer’s work. The human coder found 177 true bomb craters. The first stage of the researcher’s model identified 89 percent of the true craters, but also identified 1,142 false positives – crater-like features not caused by bombs. Much of the land covered in the study is agricultural, meaning that local farmers are at risk of encountering an unexploded bomb.


“My concerns are not about a futuristic AI that is self-aware and pursues its own goals that do not match human goals. I’m more afraid of what is already possible now. Even today, there is immense potential to lose control to an AI.”
Author and journalist Tom Hillenbrand on the dangers of using AI.


Will AI soon decide over life and death? The improvement of artificial intelligence in healthcare is leading to ethical conflicts. For example, researchers at Stanford University developed a neural network that predicts the life expectancy of patients. The system is said to have an accuracy rate of 90 percent. Although this is intended to provide better care for terminally ill patients, there is a risk that medical professionals could place too much trust in the expertise of AI. The developers of AI systems often do not provide insight into the decision-making process used by an AI. Aspire Health, which is funded by Google’s parent Alphabet, also offers a death forecast. But Aspire Health focuses on offers for outpatient palliative care, i.e. the treatment of seriously and terminally ill people. The company could therefore program its AI to classify as many people as possible as terminally ill.

Newsletter subscription

Subscribe to our free weekly newsletter for a compact overview of Artificial Intelligence topics:


More digital news briefings

Our political briefings