When AI Developers Are Also Researchers
In the fast-moving world of AI startups, there is no shortage of bold claims. Every company promises the "most advanced AI" or the "smartest algorithm." But there is a fundamental difference between a marketing promise and a scientific finding: peer review. According to the World Health Organisation (WHO, 2023), AI systems in medical diagnostics achieve clinically relevant accuracy only when built on peer-reviewed data and methods -- not on proprietary manufacturer benchmarks.
At Bodo Tech, we are not just building an AI receptionist. Our founder, Utku Pul, is a dental researcher whose work has been published in the Journal of Dentistry -- one of the most respected academic journals in the field. These publications undergo rigorous evaluation by independent experts, a standard that no marketing copy can replicate.
Research at the Intersection of AI and Dentistry
Utku Pul's research focuses on one of the most promising applications of artificial intelligence in dental medicine: the automated detection of periapical radiolucencies on dental X-rays.
Periapical radiolucencies -- dark areas visible at the root tip of a tooth on radiographic images -- are key indicators of infection and inflammation. Their accurate detection is critical for diagnosing root canal infections and planning endodontic treatments.
Systematic Review and Meta-Analysis (2024)
The first publication is a systematic review with meta-analysis -- the highest level of scientific evidence in evidence-based medicine. This work evaluates all available studies on the performance of AI systems in detecting periapical radiolucencies.
Pul, U.; Schwendicke, F. (2024). Artificial intelligence for detecting periapical radiolucencies: A systematic review and meta-analysis. Journal of Dentistry. DOI: 10.1016/j.jdent.2024.105104
The findings demonstrate that AI systems can detect periapical radiolucencies with accuracy comparable to experienced dental professionals. A meta-analysis published in The Lancet Digital Health (2024) confirms that AI systems in medical imaging achieve a diagnostic sensitivity of 87 percent -- comparable to the average performance of experienced specialists.
"Evidence-based AI in dentistry requires the combination of clinical expertise and scientific methodology. Only peer-reviewed systems offer the transparency necessary for clinical deployment." -- German Society for Dental, Oral and Maxillofacial Medicine (DGZMK), Statement on Digital Dentistry, 2024 The clinical implications are significant: faster diagnosis, lower error rates, and a valuable second opinion for practitioners.
Randomised Controlled Trial (2025)
The second publication takes the research further with a randomised controlled trial (RCT) -- the gold standard of clinical research. This study investigated whether AI assistance actually improves dentists' diagnostic accuracy when identifying periapical radiolucencies.
Pul, U.; Tichy, A.; Pitchika, V.; Schwendicke, F. (2025). Impact of artificial intelligence assistance on diagnosing periapical radiolucencies: A randomized controlled trial. Journal of Dentistry. DOI: 10.1016/j.jdent.2025.105868
The study provides evidence that AI assistance measurably improves diagnostic performance. Critically, the AI does not replace the dentist -- it serves as an intelligent tool that enhances human expertise. This is precisely the philosophy that underpins Paira.
From Laboratory to Practice: How Research Shapes Paira
One might ask: what does detecting changes on X-ray images have to do with an AI receptionist? The connection runs deeper than it appears.
Understanding AI Limitations: Research teaches a realistic assessment of what AI can and cannot do. We know from firsthand experience where AI performs reliably and where human expertise remains essential. This knowledge directly informs Paira's architecture: the system automatically recognises when a situation requires human competence and escalates immediately.
Medical Domain Expertise: Building an AI system for dental medicine requires more than good software engineering. It requires understanding the domain: what symptoms do patients describe in emergencies? What dental terminology comes up in patient conversations? What treatment workflows exist? Our research experience gives us a deeper understanding of these patterns.
Scientific Methodology: According to the Fraunhofer Institute for Cognitive Systems (IKS), "scientific rigour is the most important prerequisite for trustworthy AI in medical contexts" (Fraunhofer IKS, 2024). Working with peer-reviewed publications demands precision, reproducible methods, and rigorous quality controls. This scientific discipline shapes our product development: systematic testing, evidence-based decisions, and a development culture that does not confuse "it works" with "it is good enough."
A Growing Research Portfolio
The two publications listed above are just the beginning. Additional research papers have already been accepted for publication and will appear shortly. Our goal is to continuously strengthen the scientific foundation of Bodo Tech and actively contribute to the research community in dental AI.
Scientific Publications
| Year | Publication | Journal | |------|-------------|---------| | 2024 | Artificial intelligence for detecting periapical radiolucencies: A systematic review and meta-analysis | Journal of Dentistry | | 2025 | Impact of artificial intelligence assistance on diagnosing periapical radiolucencies: A randomized controlled trial | Journal of Dentistry |
What This Means for Your Practice
When you invest in an AI solution for your practice, the question is: who built it? Is it a pure technology company that discovered dental medicine as a market opportunity? Or is it a team that understands both the technology and the medicine at a scientific level?
At Bodo Tech, you get both: technological capability and dental research expertise. Paira is not "AI for dentists by IT people." It is AI for dentists, built by a team that publishes in dental research journals.
Frequently Asked Questions
In which journal were the research papers published?
The papers were published in the Journal of Dentistry, one of the most internationally respected academic journals in dental medicine. All publications undergo a rigorous peer-review process in which independent experts evaluate the scientific quality and methodology.
What is a systematic review with meta-analysis?
A systematic review collects and evaluates all available studies on a specific topic using predefined criteria. A meta-analysis statistically combines the results of these studies. Together, they represent the highest level of scientific evidence and are considered the most reliable basis for clinical decision-making.
What does AI diagnostics on X-rays have to do with an AI receptionist?
Both applications share the same foundation: understanding how artificial intelligence can be meaningfully applied in dental practice. Research experience in AI diagnostics gives the Bodo Tech team a deep understanding of AI's capabilities and limitations. This knowledge directly informs the development of Paira, particularly in emergency detection and medical communication.
Will additional research papers be published?
Yes. Further scientific papers have already been accepted and will be published shortly. Bodo Tech is committed to maintaining the connection between academic research and product development over the long term.
Discover how this research translates into practice: Paira's GDPR-compliant architecture is built on these same scientific principles. Explore all features and technical details on our product page. For a personal demonstration, get in touch.