2021 | Health | Promising | SDG10 | SDG17 | SDG3 | SDG9 | United Kingdom
An innovative AI diagnostic tool using saliva to predict oral health vs disease

1. General


SDG 3: Good Health and Well-being

SDG 9: Industry, Innovation and Infrastructure

SDG 10: Reduced Inequality

SDG 17: Partnerships to achieve the Goal



2. Project Details

Company or Institution

CircaGene ltd.


An innovative AI diagnostic tool using saliva to predict oral health vs disease

General description of the AI solution

The global burden of oral diseases affects 3.9 billion people worldwide – 4th most expensive disease to treat. CDC report showed that 47.2% of adults over 30 years of age have periodontal disease, which increases to over 70% for over 65-year-olds. Annual direct and indirect costs reach $442B and recent evidence shows that dental diseases can lead to serious systemic diseases such as hearth diseases, Alzheimer’s, and rheumatism in untreated. WHO states the importance of lowering the burden, but the field is lacking innovation. CircaGene develops an innovating diagnostic tool using saliva combined with AI for early detection and prevention of oral diseases. We use non-invasively collected saliva to run deep genomics analyses to devise biomarkers for oral. The genomic profile based on specific and carefully selected biomarkers (Identification of both pathogenic bacteria and their Antibiotics Resistance genes) help us detect oral diseases early enough to make them reversible. We’re able to accurately diagnose the disease and its severity and then devise the right personalised treatment (antimicrobial resistance). Our technology is at TRL 5 stage (technology demonstration) as we have recently run a successful clinical study that confirmed the 90% accuracy of our AI algorithms on patients data. From an equality perspective, dental diseases affect people from low-income countries (LICs) with limited access to dental clinics, poorer diets, and health awareness. We aim to provide this technology for free to regions of the highest need (LICs in Africa, Latin America, and Asia) together with the health awareness campaign. We have existing collaborations with Institute Pasteur Senegal (Department of Microbiology led by Dr DIEYE) and plan in the short-term to do free screening for over 100 adults and kids. We aim to start with selected LICs countries and expand the list progressively with guidance from WHO.





3. Aspects

Excellence and Scientific Quality: Please detail the improvements made by the nominee or the nominees’ team or yourself if your applying for the award, and why they have been a success.

Our technology is the first of its kind in this field. Identification and treatment of oral diseases are solely empirical, and the therapy choice is an educated guess. It means that when patients come to the dentist, it is already too late, and the disease is not reversible. We aim to detect dental diseases at an early stage where hygiene adjustment, diet improvement or treatments can reverse the disease completely. To do so, our technology unlocks the knowledge about oral and systemic health directly from saliva. Saliva is non-invasively collected and analyzed for specific bacterial genetic genes to predict the disease at an early stage or to detect it accurately including an exact understanding of pathogenesis and the right personalized treatment e.g., based on antimicrobial resistance to antibiotics. CircaGene uses genomics to build complex predictive models of health and disease. Specific focus on dental decay, gingivitis, periodontitis which are reversible if detected early. AI and mathematical modelling helped us develop a complex fingerprint profile of what we understand as oral health and each of the diseases have specific biomarkers giving a predictive score for the disease. We have run a clinical study in collaboration with a dental clinic in London where we tested our predictive algorithms and established an accuracy of 90% for our algorithm. More precisely, a list of classifiers was selected for comparing their effectiveness; Logistic Regressions demonstrated the highest accuracy with 90% on a Leave-One-Out cross validation metric using only the original z-scored vectors. Our existing collaborations with Institute Pasteur Senegal will allow us to screen 100 subjects and we’ve established a collaboration with Professor Francesco D’Aiuto (World leading expert in oral diseases) at UCL London Eastman dental Institute to expand the sampling data set and train our model even further.

Scaling of impact to SDGs: Please detail how many citizens/communities and/or researchers/businesses this has had or can have a positive impact on, including particular groups where applicable and to what extent.

CircaGene is passionate about public health as evidenced by our research collaborations with LIC partners: Institute Pasteur Senegal. Access to healthcare and dentists is the most difficult in low-income countries lacking educated staff and infrastructure. Our AI-powered diagnostic is a leapfrogging technology enabling unprivileged societies to have access to preventative Precision Oral Health.
Our technology contributes to SDG3. Good health and wellbeing, 10. Reduced Inequalities (to access Precision Health technologies), 17. Partnerships for the goals (with local researchers and Healthcare providers), 9. Industry Innovation and Infrastructure SDGs. This Saliva Health test is simple, painless, accurate and affordable. Patients can leave their saliva sample in hospitals, selected local shops and pharmacies. They only need to spit into a tube that allows for the stabilization of DNA. Each sample is sent to our collaborator labs for the genomic analysis. We plan to work closely with Institute Pasteur’s network of research labs in Africa to do our testing. This will create local qualified jobs and encourage further development of research infrastructure in the regions. Further AI analysis will be done in the cloud and shared with patients via their phone. Clear and well-explained health outcomes & information, recommendation and treatment options will be provided. We believe that using our service twice a year and following the guidance, will help reducing significantly oral health-related diseases as well as monitor emerging antibiotics resistances in the population. CircaGene will leverage the profits made from selling the same product in high income countries to provide free of charge service to the poorer communities. We intend to publish and share our findings to raise awareness.

Scaling of AI solution: Please detail what proof of concept or implementations can you show now in terms of its efficacy and how the solution can be scaled to provide a global impact ad how realistic that scaling is.

This is the first DNA-based AI-powered genomic technology dedicated to diagnosing oral health diseases. Currently, dentists diagnose diseases based on visual evaluation, where the disease is already well-established and thus irreversible, and without access to microbiome composition nor Antibiotics resistance profiles. Our technology renders oral diseases reversible as they can be detected at an earlier stage and eliminated by oral hygiene and personalized treatments. We estimate that regular (bi-annual) use of CircaGene test will prevent 80% of dental diseases such as caries, gingivitis, and periodontal diseases; subsequently reducing long-term chronic diseases enabled by poor Oral health. Our product is fully scalable, and we are already selling it in the UK (as Research Use Only). The main challenge is establishing the same levels of services in LIC free of charge, through an innovative Business model. The lowest cost per patient per year (Two separate tests) is approx. £96. The cost relates to the bare laboratory expenses for complex genomics analyses. We believe this cost can be partially offset by leveraging the profits made from the same services in high income countries. We would also seek lowering the laboratory costs by working with local providers in the Institute Pasteur's network. The remaining financing for such testing will be provided by Crypto-Philanthropy, a fully traceable blockchain-based financing of healthcare in developing countries (discussions on-going with a partner company). In terms of impact, we believe our diagnostic tool will revolutionize the Oral Health field and popularize AI genomics technologies; a powerful technology becoming cheaper as we speak and therefore more accessible to the public. Our network contains early adopters / clients who access health reports and advice through an online interface or app. All patient’s data are encrypted, and CircaGene owns a patent WO2020240167A1 on secure methods for human genomic analysis.

Ethical aspect: Please detail the way the solution addresses any of the main ethical aspects, including trustworthiness, bias, gender issues, etc.

Moral and ethical consideration for our product include 1) equal access to high-quality healthcare for both HICs and LICs, 2) data securing and 3) accurate health prediction to avoid any misdiagnosis and 4) compliant with regulations and local laws.
Ad 1) Our leapfrog-technology allows access to sophisticated healthcare to people from regions where dental clinics don't exist, there's no educated dentists/hygienists to provide dental advice. Having access to a mobile phone and the internet is enough to monitor the oral health and follow diet, hygiene, and treatment recommendation. Such diagnostic technology can be accessible to both North and South regions through innovative financing models and leveraging our profits from sales in HICs to provide free of change testing to our patients in LICs.
Ad 2) Our technology relies on sensitive personal health data. We understand the importance in protecting personal & health data hence developed and patented a secure method of accessing and processing patient data in a highly encrypted way (WO2020240167A1/GB1907358.4).
Ad 3) We have run a clinical study to test our AI algorithms (data available upon request) and can confirm the predictive accuracy of 90%. We have two ongoing collaborations with the Institute Pasteur Senegal and Prof D'Aiuto from UCL Eastman Dental Institute to perform further populational studies of 100 patients and beyond in both high income and low-income countries (UK, Senegal). Extensive clinical studies run with local population and AI-algorithm validation will avoid misdiagnosis and help us adjust the modelling for any heterogeneity amongst different populations/cohorts.
Ad 4) We work with regulatory agencies (MHRA, EMA) and advisors (OMC Medical) to meets all necessary regulations. Under new EU-IVDR regulations, our product is classified as an in-vitro diagnostic, and we started the regulatory process for ISO accreditation and CE marking.


International Research Centre
on Artificial Intelligence (IRCAI)
under the auspices of UNESCO 

Jožef Stefan Institute
Jamova cesta 39
SI-1000 Ljubljana



The designations employed and the presentation of material throughout this website do not imply the expression of any opinion whatsoever on the part of UNESCO concerning the legal status of any country, territory, city or area of its authorities, or concerning the delimitation of its frontiers or boundaries.

Design by Ana Fabjan