#4D6D88_Small Cover_March-April 2024 DRA Journal

In this exclusive Show Preview Issue, we present the IDEM Singapore 2024 Q&A Forum featuring key opinion leaders; their clinical insights covering orthodontics and dental implantology; plus a sneak peek at the products and technologies set to take center stage at the event. 

>> FlipBook Version (Available in English)

>> Mobile-Friendly Version (Available in Multiple Languages)

Click here to access Asia's first Open-Access, Multi-Language Dental Publication

AI Advances Dental Caries Detection, Study Finds

SAUDI ARABIA: Researchers from King Abdulaziz University in Saudi Arabia, Faculty of Medicine at Cairo University in Egypt, Virginia Polytechnic Institute and State University in Virginia, and the Faculty of Dentistry at the University of British Columbia in Canada have collaborated on a research paper titled “Artificial Intelligence in the Detection and Classification of Dental Caries.”

The primary objective of their study, published on August 26, 2023, is to explore the potential of artificial intelligence (AI) in the detection and classification of dental caries, commonly known as tooth decay. The research team recognised that automated detection of dental caries could significantly improve early diagnosis, streamline clinician workflows, and enhance treatment decision-making.

Read: Korean varsities advance AI-based dental diagnostic research

Leveraging Deep Learning

The research methodology involved the collection of bitewing radiographs with a resolution of 1876×1402 pixels. These radiographs were meticulously processed, segmented, and anonymized using radiographic image analysis software. The key innovation lay in the use of deep learning models, specifically supervised learning algorithms trained on semantic segmentation tasks, to identify and classify dental caries based on the modified King Abdulaziz University (KAU) classification system.


Click to Visit website of India's Leading Manufacturer of World Class Dental Materials, Exported to 90+ Countries.


 

The study yielded promising results, with the AI model achieving a mean score of 0.55 for proximal carious lesions on a 5-category segmentation assignment. Additionally, the model demonstrated a mean F1 score of 0.535, utilising a dataset comprising 554 training samples.

Read: Aoikai develops AI diagnostics system

Transforming Dental Care

The implications of this research are profound. The successful development of an accurate caries detection model has the potential to expedite the identification of dental caries, thus enabling prompt treatment and prevention measures. Furthermore, it holds the promise of optimising clinician decision-making and elevating the overall quality of patient care.

This collaborative effort between researchers in Saudi Arabia, Egypt, the United States, and Canada underscores the global significance of this breakthrough in dental care. As AI continues to make inroads into the field of dentistry, it is anticipated that such innovations will reshape dental practice and contribute to improved oral health worldwide.

Read: AI-Enhanced Dental Imaging: Precise Tooth Segmentation and Damage Prediction

The information and viewpoints presented in the above news piece or article do not necessarily reflect the official stance or policy of Dental Resource Asia or the DRA Journal. While we strive to ensure the accuracy of our content, Dental Resource Asia (DRA) or DRA Journal cannot guarantee the constant correctness, comprehensiveness, or timeliness of all the information contained within this website or journal.

Please be aware that all product details, product specifications, and data on this website or journal may be modified without prior notice in order to enhance reliability, functionality, design, or for other reasons.

The content contributed by our bloggers or authors represents their personal opinions and is not intended to defame or discredit any religion, ethnic group, club, organisation, company, individual, or any entity or individual.

Leave a Reply

Your email address will not be published. Required fields are marked *