Ilman Shazhaev, is the Co-Founder & CEO of Acoustery, a health-tech company that develops AI technology for the early recognition of respiratory diseases.

What initially attracted you to computer science and engineering?

The quantity of information available today is more extensive than ever, and AI technology — which may be very data-dependent — has made tremendous progress up to now few years. This is the reason doing research on this field is so exciting.

Without delay, I’m focused on Big Data projects. During COVID-19, I  co-founded Acoustery: a totally automated AI-powered solution for monitoring one’s health based on the evaluation of their voice, cough, and breath.

The subsequent step was to mix health research and gaming. Why? The quantity of information this industry generates is exclusive; what’s more, gamers are early adopters able to share their data and contribute to scientific progress. At the identical time, the variety of ongoing clinical trials is low, the progress is slow, and the gaming sector allows for way more dynamic data processing.

Could you elaborate on the genesis story behind Acoustery?

​​​As I discussed before, Acoustery was began in the course of the pandemic. Although business opportunities in 2020 were relatively limited, I used to be staying in Dubai, certainly one of the few locations where a project could operate without super strict limitations.

My co-founder Dr.Dmitry Mikhaylov, a professor on the National University of Singapore, and I began on a recent challenge: early-stage detection of COVID-19. On the time, UAE was massively exploring early diagnosis technologies and largely supported AI projects.

Because of this, we got access to among the best testing facilities within the UAE: Sheikh Zayed military hospital, where we had data from a whole lot of COVID-19 patients to coach our AI engine on.

At the subsequent stage, tests showed our technology was very accurate and had great potential. Researchers published their leads to the highest rate journals in Japan and USA, and our testing method was utilized in several Asian countries during pandemics as an emergency tool.

When COVID-19 was over, we focused on detecting asthma using the identical approach. Sharjah University, which is currently leading in UAE’s research, ground-approved these tests.

For COVID-19 how accurate is this method in comparison with PCR, LFT, and antibody tests?

The positive predictive value of Acoustery within the context of community-wide screening for COVID-19 is comparatively high (81%) in comparison with Xpert MTB/RIF, a recent test that’s revolutionizing tuberculosis detection and control by contributing to the rapid diagnosis of the disease (61%) and PCR throat swabs (71%).

Our findings have shown that the software developed by Acoustery will be used as a primary non-laboratory screening tool to detect cases of COVID-19 and route patients to laboratories for PCR testing.

Could you tell us more concerning the machine learning used to coach the AI?

We assumed that to get an accurate detection rate of COVID-19, we could train convolutional and recurrent networks to diagnose the disease by analyzing the spectrograms of cough and breath of patients. A spectrogram is a visible way of representing the signal strength at various frequencies. Quite a lot of medical studies showed significant differences between the cough of patients who had COVID and those that didn’t, so we trained our AI engine to acknowledge such differences.

Acoustery’s developments will be used to diagnose Alzheimer’s, which is often perceived as a neurological disorder. How exactly does it work?

Our study explores how speech measures could also be linked to language profiles in participants with Alzheimer’s disease (AD) and the way these profiles could distinguish AD from changes related to normal aging. To attain this, our AI analyzes easy sentences pronounced by older adults with and without AD, from the share and variety of voice breaks to shimmer (amplitude perturbation quotient) and noise-to-harmonics ratio. The accuracy of this evaluation reaches 90%.

In a while, we used the identical approach in Farcana Labs – a enterprise focused on collecting Big Data generated by gamers to research disease progression, especially with mental disorders.

What other diseases will be diagnosed using this method?

Asthma is our key priority now. Tuberculosis is one other focus, in addition to chronic obstructive pulmonary disease (COPD), pulmonary fibrosis, pneumonia, and lung cancer.

How large are the training data sets for these use cases?

We’ve 1000’s of cough recordings in our database collected in the course of the last 4 years.

What’s your vision for the long run of medical diagnosis across the board?

The info collected by personal devices will play a big role in diagnosing diseases at an early stage and stopping pandemics. Even our mobile phones have multiple sensors: a microphone is just certainly one of those. Accelerometers that may analyze motor skills and detect quite a few diseases are one other.

Although these technologies shouldn’t be the one source for diagnosing, they’ll significantly help predict and forestall the spread of highly infectious respiratory diseases — and,  consequently, recent pandemics. Acoustery can be utilized in developing countries where access to PCR testing is restricted.

You appear to have multiple projects on the go; what are another exciting use cases that you just see for AI?

The AI space is exclusive. As AI researchers, we concentrate on niches that generate big data, which is essential for any AI research. We want numerous patients to compile quality datasets, so we have now a couple of pieces of research moving into parallel and are exploring several business verticals.

We see gaming as an area where an enormous amount of information is generated. Today, people play numerous video games, which is a beneficial source of information for health research. Collecting data from personal devices and wearables is one other vector with significant potential.

All in all, it’s exciting to be exploring this technology now, and I feel it has way more potential still to be harnessed across other sectors.