How are they used? Satellites are often the first tool to detect wildfires in remote areas, using cameras, lidar, and infrared sensors to capture how actively they are burning, their precise location, and the direction of the resulting smoke. NASA has a fleet of satellites it uses to observe Earth, with equipment in geostationary orbit providing imagery every five to 15 minutes. The satellites take a "snapshot" as they pass over the region, capturing an area of up to a kilometer at a time. An algorithm uses infrared data to detect if there are any fires burning, examining each kilometer chunk and labeling it as fire, non-fire, cloud, water, missing data, or unknown. This data is sent to fire management authorities worldwide, and used both for operations and for mapping the scale and type of the damage once the fires have burned out. 

See for yourself: This image of the fires shows the areas that have been affected since December. It was created by photographer Anthony Hearsey using data from NASA’s Fire Information for Resource Management System (FIRMS), covering the month up to January 5, 2020. FIRMS sends data to those who need it within three hours of its being captured by two NASA satellites.

More bad news: There’s been some temporary respite in southeast Australia over the weekend, thanks to rain and cooler temperatures. However, that brings its own hazards in the form of dangerous levels of smoke, and the weather is likely to warm up again toward the end of the week. 

Read next: A team of researchers are trying to understand the science behind the world’s worst wildfires.

Share

Link

Author

Charlotte Jee

ImageNASA

The news: DeepMind and Google Health have developed a new AI system to help doctors detect breast cancer early. The researchers trained an algorithm on mammogram images from female patients in the US and UK, and it performed better than human radiologists. The results were published in Nature on Wednesday.

A tragedy of errors: Breast cancer is the most common cancer for women globally, and their second leading cause of death. Though early detection and treatment can improve a patient’s prognosis, screening tests have high rates of error. About 1 in 5 screenings fail to find breast cancer even when it’s present, also known as a false negative; 50% of women who receive annual mammograms also get at least one false alarm over a 10-year period, known as a false positive.

The results: In tests, the AI system decreased both types of error. For US patients, it reduced false negatives and positives by 9.4% and 5.7%, respectively; for UK patients it reduced them by 2.7% and 1.2%. In a separate experiment, the researchers tested the system’s ability to generalize: they trained the model using only mammograms from UK patients, and then evaluated its performance on US patients. The system still outperformed human radiologists, reducing false negatives and positives by 8.1% and 3.5%.

Why it matters: The system’s ability to generalize in this way has promising implications. It shows that it may be possible to overcome one of the biggest challenges facing AI adoption in health care: the need for ever more data to cover a representative patient population. But such results should also be interpreted with caution. Relatively speaking, the US and UK have quite similar populations. The system likely would not generalize as well to other parts of the world.

Related work: Last October, NYU researchers published a similar study, demonstrating an AI system for breast cancer screenings on par with human radiologists. The primary differences, however, were that it only used mammograms from US patients, and it compared the system’s performance with human expert diagnoses conducted in an artificial lab environment. Google and DeepMind instead compared performance with real-world diagnoses.

Human and machine: Ultimately, both studies conclude that such AI breast cancer screenings should be used in tandem with human radiologists. The combination achieves the most accurate diagnostic results but still reduces the workload on human radiologists, which would help free up their time to focus more on patient care.

Share

Link

TaggedTech and Health

Author

Karen Hao

In 2019, these attacks wreaked havoc around the globe, earned criminals vast sums, and even occasionally provided a weapon for government hackers. This marked the fifth straight year of growth, with national and local governments and public institutions increasingly becoming targets.

The money: The potential cost of ransomware in the United States last year was over $7.5 billion, according to a recent report from the cybersecurity firm Emisoft that attempted to estimate the impact of a very opaque set of incidents. 

The victims: Emisoft tallied up 113 governments and agencies, 764 health-care providers, and up to 1,233 individual schools affected by ransomware in America. Big cities including Baltimore and New Orleans were both struck by ransomware attacks last year.

The why: One root cause, according to an October 2019 report from the State Auditor of Mississippi, is a “disregard for cybersecurity in state government.” Others agree: Research from the University of Maryland published earlier in the year concluded with admirable directness “that most American local governments do a poor job practicing cybersecurity.”

This isn’t a problem just for small towns and their ill-equipped agencies. Last month, a US Coast Guard facility was forced offline for over 30 hours when ransomware hit the base's cameras, access systems, and critical monitoring systems, the BBC reported.

Share

Link

Author

Patrick Howell O'Neill

Image"Cryptolocker ransomware" by Christiaan Colen is licensed under CC BY-SA 2.0