Will artificial intelligence recognize COVID-19 and other diseases?

March 18, 2021

Artificial intelligence (AI), step by step is making its way into the healthcare field. As a disease recognizing tool, AI could help people follow and monitor their health without actually going to medical care centres. It could help improve the time and efficiency of disease recognition. But the question is, how can artificial intelligence find and recognize diseases of the human body?

The finalists of Vilnius Gediminas technical university's (VILNIUS TECH) makers academy "Makeademy" and the inventors of the "Health Check Device" ("HCD"), Jonas Rudis and Ignas Kuncevičius – found the answer to this question. The author of the "Health Check Device" idea, Jonas Rudis, says that the "HCD" is designed to recognize various human diseases, including COVID-19. For predicting sickness, the device collects mechanical sounds of the human body and uses an artificial intelligence sound recognition system. 

The idea of the "HCD" product came to J. Rudis and I. Kuncevičius during their studies at the Vilnius Academy of Arts (VDA), studying the consequences of the COVID-19 virus and the possibilities of facilitating them.

Recognition of diseases – through the sounds of the human body

"Attached to specific parts of the human body, the "HCD" device can identify and record sounds emitted by the human organism. The received information is then processed and classified by AI. AI uses specific algorithms in order to identify diseases according to the available collected material," J. Rudis, the creator of the idea, says. 

J. Rudis also states, that "the collected material about a person becomes additional data that helps AI improve and recognize more symptoms. Yet, for security reasons, the user profile is anonymous".

The device can become a great help in the field of healthcare. According to "HCD" co-founder I. Kuncevičius, "HCD" can be applied in various ways: from preventive health care at home, remote patient care to health monitoring of professional athletes.

Last year the young creators with the product's idea decided to step further and to participate in the technical creativity program "Makeademy", which was organized by VILNIUS TECH Creativity and Innovation Centre "LinkMenų fabrikas" and telecommunication company "Tele2". 

The future of artificial intelligence in medicine

"We are going to see AI diagnosing diseases everywhere in a few years," Lucas Lopez, the CTO and co-founder of "Volatile AI" and one of the mentors of the "Makeademy" program, says. 

"HCD is developing a hardware device aiming to improve the performance of predictions, which definitely has great potential. However, there is a lot of competence in the field now, and data science is what will make a difference. They need to play their cards well to position in the market," L. Lopez adds.

Still, Lucas Lopez predicts that the "HCD" team in the future may face several challenges that could catch up with any start-up working in the AI field.

"I see two big challenges. Firstly, they need to prove that it works; that will take a long time due to the long-term development nature of diseases and the strict medical regulations in medicine, which means a need for capital or source of revenue that will allow them to survive until then. Secondly, the adoption of the technology will be slow by both experts and users; AI can prove that it works statistically but gives no explanations of how or why, and that is very difficult to accept by people," L. Lopez says. 

The research 

During the "Makeademy" project, the "HCD" creators participated in various rapid prototyping workshops and courses. They developed the design of their prototype and delved into the possibilities of using artificial intelligence in medicine.

"After the consultations with the specialists developing the AI, we learned that at the stage of data collection, a simple telephone microphone is used to collect the sounds of the human body. When using such a microphone, a lot of extraneous information is collected, and relatively few sounds emitted by the human body are recorded," J. Rudis says. 

Thus, the idea of creating a device that can record the sounds of the human body without extraneous background noise was born. 

Firstly, the "HCD" developers focused only on recognizing the COVID-19 disease. Yet, a variety of new scientific articles have been published. By collaborating with several companies developing artificial intelligence for disease recognition, developers have learned that it is possible to identify a much broader range of sounds than just the ones produced by COVID-19. Consequently, the developers decided to create a device recognizing more human diseases. 

"At the beginning of the project, we collaborated with the "VoiceMed" team from Luxembourg. They gave us the basic knowledge of product development and advised us how to apply the technology they are developing in our "HCD" project," J. Rudis says.

"Later on, we started working with other AI development companies – "NeuroLex" (United States) and "Corti" (Denmark). These companies help us to understand which diseases of the human body can be detected. Also, we discussed what problems AI-based companies can face and what legal requirements arise when recording body sounds," J. Rudis adds. 

"The development of the "HCD" project became a real challenge for both of us", the designer I. Kuncevičius admits. 

"In addition to design and construction work, it was necessary to delve into the psychology and technology of the prospective user, gain knowledge of the characteristics, strategies and regulations of the medical business," I. Kuncevičius says. 

Upcoming plans

The "HCD" team shared second place in the "Makeademy" project together with team "Algy". The "Algy" team is developing an innovative micro-algae aquarium that purifies the air from carbon dioxide. Both teams won 1,500 euros each.

The "HCD" founders admit that they are going to invest their whole prize back into the project. J. Rudis and I. Kuncevičius also state that the "Makeademy" program has helped them find helpful contacts and provided them with all the support needed to start their project. 

"Programs like "Makeademy" are essential for society. They offer highly skilled and motivated individuals to apply their knowledge in new ways and make impact. Very often the skills of workers are not exploited to their full potential in corporates, and showing people there is an alternative way is like removing a blindfold," Lucas Lopez, the CTO and co-founder of "Volatile AI" adds.
 

Related news

€75,000 Funding Opportunity for Women-Led Deep Tech Startups
€75,000 Funding Opportunity for Women-Led Deep Tech Startups
Women-led deep tech startups are invited to apply for the European Union-funded "Women TechEU 2" program. This initiative aims to strengthen women's entrepreneurship in the technology sector, foster innovation, and help promising startups prepare for further growth and attract additional funding. Selected participants will receive a €75,000 grant and the opportunity to take part in a six-month tailored business development program. During this period, startups will be provided with mentoring, consulting, and training services designed to accelerate business growth and enhance competitiveness in the international market. The total budget for the "Women TechEU 2" program is €12 million. The plan is to fund 160 women-led startups from across Europe. Application Process The application process for the new call consists of two stages. First, applicants must submit an application for the Eligibility Strand to verify whether the startup meets the program's requirements. Applicants deemed eligible will then be invited to submit a Full Proposal. The eligibility assessment stage began on June 1, 2026. Applications are accepted on a rolling basis, with evaluations conducted weekly – the weekly submission deadline is every Tuesday by 18:00 (17:00 CEST). The first deadline is June 30, 2026, and the final deadline is July 13, 2027. Applicants will receive their eligibility assessment results no later than one week after the respective submission deadline. Who Can Apply? The program is designed for women-led, early-stage deep tech startups. To be eligible for funding, the following conditions must be met: At least one founder or co-founder of the startup is a woman; A woman holds a top-level executive position (CEO, CTO, or equivalent); Women hold at least 25% of the company's shares. Program Objectives The "Women TechEU 2" program aims to: Promote women's entrepreneurship in the deep tech sector; Help innovative startups scale up and attract additional European Union funding; Strengthen European innovation ecosystems; Support the development and commercialization of new technologies. Potential applicants are encouraged to review the application guidelines in advance and avoid waiting until the final deadline to submit their proposals. More information can be found here: https://womentecheurope.eu/active-calls/
More
New doctoral dissertation
New doctoral dissertation
VILNIUS TECH Library invites you to follow the published new dissertations. The dissertation „Performance investigation of a hybrid car engine fuelled with gasoline and gaseous mixtures“ („Benziną ir dujų mišinius naudojančio hibridinio automobilio variklio efektyvumo tyrimas“) prepared at VILNIUS TECH by Tadas Vipartas. The dissertation was prepared in 2021–2026. Scientific consultant – Prof. Dr Alfredas Rimkus. The dissertation was defended at the public meeting of the Dissertation Defence Council of the Scientific Field of Transport Engineering in the Aula Doctoralis Meeting Hall of Vilnius Gediminas Technical University at 9 a.m. on 12 June 2026. This dissertation investigates the use of alternative fuels (natural gas and hydrogen) to increase the efficiency of a spark-ignition engine. The impact of different fuels and engine control algorithms on the combustion process and on energy and ecological indicators was determined and evaluated by analysing the emerging technological constraints within the context of a power-split (series-parallel) hybrid powertrain. The dissertation presents a review of scientific literature, analysing the directions for internal combustion engine improvement, the properties of gaseous fuels and the challenges of their application, along with the operating principles of automotive hybrid powertrains. Bench tests were conducted to investigate the effect of late intake valve closing timing on an engine operating on natural gas, and the influence of hydrogen additives on the combustion process and knock control. The numerical analysis of the engine’s combustion process was performed using AVL BOOST™ software, while the energy and ecological indicators of the hybrid vehicle were evaluated through experimental research and numerical simulation in the AVL CRUISE™ software. The following main results were obtained in the dissertation: retarding the intake valve closing timing increased the brake thermal efficiency and NOx emissions while reducing carbon dioxide emissions when the engine operates on natural gas. It was determined that a hydrogen additive in the fuel improves the engine’s energy indicators, but increases nitrogen oxides emissions and the risk of engine knock. Engine knock is effectively managed by retarding the ignition advance angle. Numerical simulation results confirmed that these trends persist during the Worldwide Harmonized Light-duty Vehicles Test Cycle: the use of hydrogen reduces fuel consumption and carbon dioxide emissions, but increases nitrogen oxides emissions. The dissertation results revealed the potential of these technologies and strategies for their application. The obtained data can be applied in the development and selection of advanced engine control algorithms and in the formulation of technologically sound environmental standards. Nine scientific articles have been published on the topic of the dissertation: six in scientific journals indexed in the Clarivate Analytics Web of Science database with an impact factor, one in a scientific journal indexed in the Clarivate Analytics Web of Science database without an impact factor, one in conference proceedings indexed in the Clarivate Analytics Web of Science Conference Proceedings Citation Index, and one in peer-reviewed conference proceedings not indexed in international databases. The research results were presented at three scientific conferences in Lithuania and Poland. Doctoral dissertation readers can search via VILNIUS TECH Virtual Library.
More