The expansive nature of Artificial Intelligence (AI) is allowing healthcare professionals to explore its usage in their specialised field of work. One such field in dentistry is Periodontology which deals with prevention and treatment of gum diseases.

Gum diseases are a natural occurrence in an aging population, however, the education and adherence rates of people to gum treatment is low. This is because it might not be seen as an immediate need by a patient, especially in the early stages, vs the cost of treatment. In addition, some patients are biased towards diagnosis of other conditions – such as cavity fillings, root canal treatments, implants etc, – assuming that these have more detrimental impact to one’s oral health, compared to gum related problems.

Practically, gum disease presents itself in many ways. Some examples are when patients notice that their teeth seem to be ‘drifting’ from their previous location, as they age, often complained as gaps forming between their teeth, which didn’t exist before.

The most common and universally known symptom of gum disease is bleeding during brushing. All these symptoms, and more, account to gum disease which can best be controlled when captured in a timely manner.

What is Periodontology?

Periodontology is a branch of dentistry that deals with the prevention, diagnosis and treatment of diseases affecting the hard and soft tissues that support and protect the gums and teeth. Periodontal diseases present as gingivitis characterized by gum inflammation and bleeding without bone involvement, or periodontitis, where the bone supporting the teeth is destroyed. Without timely treatment, these conditions can lead to tooth loss.

In this article, we will explore how AI has transformed periodontology, enhancing the precision, efficiency, and accuracy of periodontal diagnosis and patient education.

Detection of periodontal disease

Traditionally, detecting periodontal diseases has meant relying on a mix of clinical exams and diagnostic tools. While these methods have worked, they can be time-consuming and often depends on the examiner’s consistency (subjective in nature).

The use of AI algorithms for the automatic detection of periodontal diseases using clinical images, such as intraoral photographs and radiographs, is showing promise. These algorithms can classify periodontal bone loss from panoramic radiographs and detect gingival bleeding from intraoral images.

AI-based systems have the potential to assist clinicians in the early detection and diagnosis of periodontal diseases.

Periodontal risk assessment (PRA)

Over the years, clinicians have used multiple methods for periodontal risk assessment including clinical examinations (for probing pocket depths and assessing gingival health), radiographic analysis (for evaluating bone loss via X-rays), reviewing medical and dental history (identifying risk factors like smoking and diabetes), and sometimes microbiological testing and salivary diagnostics to detect pathogenic bacteria and biomarkers.

AI tools can integrate multiple risk factors, including demographic data, medical history, and clinical parameters, to predict the likelihood of future periodontal complications.  These tools can reveal associations between patient demographics and disease-specific factors in periodontal disease. Additionally, the screening analysis method using machine learning and images can be applied to diagnose and screen for other systemic diseases.

AI provides comprehensive assessment, leading to faster and early detection of gum disease.

Assessment of periodontal bone level

Calculating radiographic bone loss (RBL) can be complicated, time-consuming, and subjective to the examiner. To assess periodontal bone levels, several techniques are being used including clinical probing to measure pocket depths, radiographic analysis with X-rays to visualize bone levels, visual inspection of gums and supporting structures, reviewing patient history for risk factors, and comparing intraoral photographs over time to monitor changes.

AI algorithms are now streamlining the process of identifying radiographic bone loss (RBL) and predicting the likelihood of tooth loss and periodontal disease. These systems automate the detection of subtle changes in bone density and structure using various imaging modalities such as panoramic radiographs, periapical radiographs, and cone-beam computed tomography (CBCT). By providing clinicians with quantitative data, AI aids in more accurate diagnosis and treatment planning. This approach reduces reliance on subjective interpretation, improving efficiency and consistency in assessing periodontal health and ultimately enhancing patient care outcomes.

Artificial olfaction for detection of bad breath

Relying on a person to judge your breath is a bit awkward and subjective. Dental AI today has come up with breath analyzers, designed to replace traditional, often inconsistent, sniff tests.

Artificial olfaction, or the electronic nose, is a game-changer for breath analysis. This is a non-invasive technique that evaluates the full spectrum of exhaled volatile compounds using an array of non-selective sensors.

Also known as an electronic nose, it’s a combination of mammalian olfaction and AI which identifies specific patterns of smell and is used as a reference for future identification. The sensor array is programmed into two subsets: the bottom panel, which has a higher affinity for volatile sulphur compounds (VSCs), and the top panel, which has a higher affinity for non-sulphuric volatile organic compounds.

This system, primarily based on nanomaterials, semi-selectively and collectively assesses the composition of exhaled breath. It uses analysis software and a database of breath patterns processed through a pattern-recognition application. A decision tree classifier then determines whether the subject suffers from oral or extra-oral halitosis and, in the latter case, can also associate the findings with different systemic diseases.

The electronic nose can pinpoint potential systemic diseases linked to a person’s breath

Classification of periodontal diseases

AI models have been instrumental in classifying periodontal diseases, differentiating between chronic and aggressive periodontitis, and accurately distinguishing inflamed gingiva from healthy tissue. Their capability to classify and differentiate diseases has shown great promise. Early diagnosis facilitated by AI using intraoral imaging devices provides significant advantages for both dentists and patients, enabling prompt intervention.

AI enhances diagnostic precision, potentially improving long-term oral health outcomes and patient satisfaction.

Conclusion

Conventional methods, although effective, can be time-consuming and require a high level of clinical expertise to ensure accurate diagnosis. While AI cannot replace clinicians, it serves as a powerful tool that enhances the precision, efficiency, and overall effectiveness of periodontal care. By integrating AI-driven monitoring systems into routine clinical practice, clinicians can identify early signs of disease recurrence and intervene promptly to prevent complications. AI has the potential to improve patient outcomes and shape the future of dental healthcare by enabling more accurate diagnoses and tailored treatment plans, ultimately benefiting the patients.

Early diagnosis is a win-win for both the dentists and patient. On one hand it saves the patient from complex and painful periodontal procedures later, and positions the dentist as a trusted periodontist who prioritizes prevention over cure.

The Global Dentists’ Pool

The Global Dentists’ Pool is an opportunity for both dental device/software companies and dentists to collaborate more closely together.

For dental companies

With more than a 100 dentists spread across 21 countries, this Pool supports dental companies by helping them

(1) either to validate their design ideas by running surveys and/or

(2) hire premium dental advisors to support in various clinical, regulatory, marketing and sales projects.

These are all dentists with specialized medtech expertise in different domains. The core focus is to avoid hiring multiple ‘non-dental trained’ freelancers and consultants and hire premium dental advisors instead who are the perfect substitute for multiple resources in the form of one highly specialized advisor.

For dentists

The Global Dentists’ Pool provides dentists with opportunities to participate in dental projects without disrupting their clinical practice. The amount of hours of engagement is flexible and dentists can upskill themselves by taking the “premium dental advisor program” which teaches how devices are made with medical device industry fundamentals.

If you would like to engage with us to support your project or advance your dental career, then reach out to us at contactus@namnrpro.org.

Author: Dr Rija Asghar

Author Profile: Global Dentist Pool member Dr. Rija Asghar has studied dentistry and graduated in 2019 with a distinction in Community Dentistry from Institute of Dentistry, CMH Lahore Medical College, Pakistan.

As an experienced clinician and as a Dental Machine Learning Annotationist, she’s currently working as a demonstrator in the department of Science of Dental Materials at IOD CMH LMC.

Dr Rija is also a Tongue Tie and Lip Tie Specialist having trained under the leading experts in the field and is aspiring to be a paediatric consultant in the future.

In her leisure time, Rija is a sports enthusiast with a true passion for tennis. She enjoys landscape and street photography & recently launched a perfume brand – from naming the perfumes to designing labels, mixing the oils to packaging, loves every aspect of it.