Pfizer set to acquire COVID-19 smartphone detection app for $100m

By Rachel Arthur

- Last updated on GMT

Pic:getty/danielallan
Pic:getty/danielallan

Related tags Pfizer Diagnostics Australia Artificial intelligence machine learning

Pfizer is set to acquire Australia’s ResApp - a company developing a smartphone app to detect COVID-19 and other respiratory diseases based on the user’s cough - for $100m AUD ($74m USD).

ResApp is currently testing the technology for detecting COVID-19 in clinical trials: but says the tech could be used to diagnose other respiratory disease including pneumonia, bronchiolitis, bronchitis, chronic obstructive pulmonary disease and asthma.

Use of machine learning and algorithms to diagnose disease

Brisbane-based ResApp was founded in September 2014 to commercialize technology developed by Associate Professor Udantha Abeyratne that uses sound to diagnose respiratory disease, and has since received funding from the Bill and Melinda Gates Foundation, the University of Queensland and UniQuest. It listed on the Australian Stock Exchange in July 2015.

It is developing ‘easy-to-use, affordable, clinically-validated and regulatory-cleared diagnostic tools that only require a smartphone’.

The instant screening test uses cough sounds to indicate the presence of infection, similar to the use of a doctor’s stethoscope. However, ResApp picks up ‘significantly more information than the sounds picked up by a stethoscope’ as well as removing the need for human interpretation. 

The machine learning approach has developed algorithms to diagnose disease from cough and respiratory sounds: drawing on a database of signatures from known clinical diagnoses.

In March, the company announced the tech accurately detected COVID-19 in 92% of people with infection​, based on a trial of 741 patients (of whom 446 were positive), ‘exceeding the real-world measured sensitivity of rapid antigen testing’.

ResApp has three clinical trials currently recruiting patients in the US and India to collect cough sounds, symptoms and validated COVID-19 status through PCR testing.

“These studies will generate the high quality data, backed by accurate ground truth labelling through PCR testing, that is needed to train a robust machine learning algorithm," ​says the company. "The studies are also collecting PCR test cycle thresholds, allowing evaluation of algorithm sensitivity to viral load.”

By using a large dataset of clinically-labelled non-COVID-19 cough samples, including coughs from patients with non-COVID-19 lower respiratory tract infections, ResApp will be able to ensure that any COVID-19 screening test developed accurately identifies COVID-19 and not other respiratory diseases.

Pfizer and ResApp will also enter into a Research & Development Licence Agreement for COVID-19, with a $3m AUD up-front licence fee, and up to $1m AUD in milestone payments based on clinical trial recruitment.

“This proposed acquisition and research collaboration add to our growing digital capabilities and bolster our efforts to pave a new era for digital health,”​ said Lidia Fonseca, Chief Digital and Technology Officer, Pfizer.

Gathering real-world evidence

ResApp’s regulatory-approved and clinically validated products include ResAppDx, a smartphone-based acute respiratory disease diagnostic test for use in telehealth, emergency department and primary care settings; and SleepCheck, a smartphone application which allows consumers to self-assess their risk of sleep apnoea. Both products are CE Marked in Europe and TGA approved in Australia. 

ResApp sees the potential of the app in aiding doctors diagnose respiratory diseases remotely, without the need for patients to visit a clinic. On January the app went live on Australian online telehealth platform Doctors on Demand.

ResApp has also signed a one-year agreement with the Dartford and Gravesham National Health Service Trust in the UK to pilot its app across four hospitals: with data and learnings set to inform opportunities across Europe.

Related topics Bio Developments

Related news

Show more