Breaking News

Alberta tech company calls for machine learning experts to help develop COVID-19 screening app

An Alberta-based technology company is reaching out to universities and machine learning experts around the world to contribute to its COVID 19 project, an app that serves as an earlier detection process for the ongoing epidemic.


The app received contributions from Western Economic Diversification Canada, Kinkaid Enterprises Inc., Universe Machine Corporation, and various anonymous donors.

To bring the project to the ground, Pleasant Solutions CEO Thomas Stachura told IT World Canada in an email.

In addition to the pleasant solution, Carademic has a number of projects under its belt, including the Paranoid project, which aims to turn off the keen ears of smart homes. “This is a company that has been around for a dozen years and whose software NASA relies on to manage employee passwords. We are also employed at Buckingham Palace, the Department of Energy.

We recently developed a line of digital privacy devices and launched it earlier this year. "We have an extensive history in the technology industry with most of our business in the B2B sector," Stachura wrote in the email.

Along with academics, the evaluation process will measure various user data points, including their oxygen level, pupil dilation, respiratory rate, acoustics, and volume, which are manufactured on most smartphones.

Blood oxygen levels will be measured by fingerprint scanning, camera dilation of the pupil, and respiratory, acoustic, and volume rates through a microphone.

This data will be loaded and incorporated into the machine learning model to create the possibility: is the user a potential candidate for self-isolation and / or definitive medical testing? There is no timeline for app availability, but anyone can download the Caridemic app on a smartphone. Cardiac is stressful instead of medical examination.

The application to aid early detection of COVID-19 will provide a valuable supplement, even in the case of mild or asymptomatic infection, within true positives versus false positives.

In symptomatic cases, the device can generate greater precision and more useful data than most medical personnel can achieve with a simple visual evaluation.

The company says the app uses its extensive data collection list based on the complexity of predicting early symptoms of the disease.

Therefore, different machine learning and data analysis teams are needed around the world to effectively deal with different algorithms for different insights. "Based on the collected key data, each team will be able to create a" derivative "that is a view of data or metadata.

Within the same machine learning team or on different teams, depending on the other We can derive the series. We need at least 10 volunteer machine learning teams around the world. We are actively helping to help machine learning experts. To participate in this award, an additional one is also available. Linked as incentives.

The company says that when it comes to finding correlations of Data. The experience required for this is technical rather than medical. "We need positivity confirmed by the algorithm.

For example, the 'acoustic signature' of pneumonia is more a biotechnology issue than a therapy. The website has a list of references that can be found here.

The company is sensitized to the initial machine learning phase of the project, volunteering for both infected and healthy (16 years or older), to initiate a global call to contribute to the data plan. In a completely anonymous procedure, participants will complete a short survey and then take a cardiovascular series of measurements.

The company says more than 10,000 volunteers will be needed to contribute the data for cardiology to be effective. Questions and data collection points can be viewed here.

To ensure data privacy, the company says it will not collect volunteer names or GPS locations, phone IP addresses, serial numbers, or operating system IDs, and the information will only be used for the development of this application.

No comments