BiofilmQuant: a computer-assisted tool for dental biofilm quantification

Area of Science:

• Biomedical Engineering
• Dental Imaging
• Computational Biology

Background:

• Dental biofilm quantification is crucial for oral health assessment.
• Current methods are either manual, time-consuming, and prone to bias, or automated but lack clinical adaptability.
• Existing automated techniques struggle to integrate expert input for correcting misclassifications.

Purpose of the Study:

• Introduce BiofilmQuant, a novel semi-automated software for dental biofilm quantification.
• Improve the accuracy, efficiency, and usability of biofilm analysis in quantitative light-induced fluorescence (QLF) imaging.
• Provide clinicians with a tool that combines automated segmentation with user-friendly manual correction capabilities.

Main Methods:

• Developed a semi-automated software tool, BiofilmQuant, for analyzing quantitative light-induced fluorescence (QLF) images.
• Employed a robust statistical modeling approach for automatic image segmentation into background, biofilm, and tooth substratum classes.
• Integrated a single-click interface for clinicians to correct any segmentation errors.

Main Results:

• Achieved high consistency and precision in segmenting over 200 QLF dental scans.
• Demonstrated the software's ability to accurately quantify dental biofilm.
• Validated the effectiveness of the semi-automated approach in handling complex QLF images.

Conclusions:

• BiofilmQuant offers a significant advancement over existing methods for dental biofilm quantification.
• The software provides a user-friendly and reliable tool for both automated analysis and expert-guided correction.
• Enables longitudinal quantitative analysis of biofilm across full sets of teeth, enhancing clinical utility.