Isolation or ICU? Indian Institute of Science’s test can tell

Unlike RT-PCR test, this model can tell doctors severity of the disease; Norwegian university collaborated with scientists in Bengaluru for it

Indian Institute of Science (IISc) researchers in collaboration with their counterparts from the University of Agder, Norway have developed a lightweight, mobile-friendly, efficient, artificial intelligence model for detection of covid-19 using lung ultrasound images.

According to its developers, the artificial intelligence model known as Mini-COVIDNET, can be used in mobile or in embedded applications, making them universally appealing, especially in the point-of-care setting. Point-of-care testing is also known as bedside testing.

Phaneendra K Yalavarthy, associate professor at the Department of Computational and Data Sciences told Bangalore Mirror, “Mini-COVIDNET is an efficient artificial intelligence model/algorithm, both in terms of number of parameters as well as memory, for the detection of covid-19 using lung ultrasound images to provide performance on par or better than a human expert. These models can be used in mobile or in embedded applications making them universally appealing, especially in the point-of-care setting.” Mini-COVIDNET compared to the RT-PCR testing method is said to reveal the severity of the disease in a much better way. It can also distinguish which patients need ICU admission and who can be managed under home care.

  • “The current gold standard test for covid-19 is the RT-PCR test and it does not reveal the severity of the disease, especially who must be admitted to the ICU and who can be treated at home. Our Mini-COVIDNET was shown to be sensitive to the damage to the pleural surface of the lung, which has been proven to have prognostic value, commonly observed in intensive care units with patients admitted and deceased due to covid-19,” Yalavarthy added.
  • Advertisement

The team which has developed the Mini-COVIDNET model said that they had been working on it since August of 2020, when covid-19 was its peak and when there was a severe shortage of ICU beds.

“We want to develop a model to assist clinicians to know who should be admitted in ICU and who should go under home care,” Yalavarthy said.

He also added that, unlike Mini-COVIDNET, the current artificial intelligence models for detection of covid-19 using lung ultrasound or Chest Computed Tomography images require specialised hardware to deploy them in clinics. “These (current artificial intelligence models) cannot be deployed in commonly utilized mobile platforms in point-of-care testing. With Mini-COVIDNET, we developed an easy-to-use model for the detection of covid-19 using lung ultrasound images,” he added.

IISc’s innovations for covid-19

  • India’s first ICMR approved Mobile Infection Testing and Reporting (MITR) Labs.
  • Contact tracing app GoCoronaGo.
  • Coswara – the sound-based covid-19 diagnostic tool.
  • PCR-free SARS-CoV-2 RNA detection test kit / affordable RT-PCR Test kit.
  • Mass spectrometers to study the virus
  • Low-cost ventilators.
  • Drones for disinfection of public places.
  • N95 masks/disinfectant devices.
  • Robot nurse Asha

Source link


Sagar Biswas

Leave a Reply

Close Bitnami banner