Red fluorescence imaging for dental plaque detection and quantification: pilot study

Zhao Liu, Juliana Gomez, Soniya Khan, Debbie Peru, Roger Ellwood

Research output: Contribution to journalArticlepeer-review

240 Downloads (Pure)


The red fluorescence of dental plaque originating from porphyrins in oral bacteria may allow visualization, detection, and scoring of plaque without disclosing agents. Two studies were conducted. The first included 24 healthy participants who abstained from oral hygiene for 24 h. Dental plaque was collected from tooth surfaces, and a 10% solution was prepared. These were scanned by a molecular spectrometer to identify the optimum excitation and emission wavelengths of plaque for developing a red fluorescence imaging system. Fourteen healthy subjects completed the second study. After a washout period (1 week), participants had a prophylaxis at baseline and abstained from oral hygiene during the study. They were monitored using the fluorescence imaging system at baseline, 24 h, and 48 h. A dentist clinically assessed plaque after disclosing and on red fluorescence images. Three descriptors were extracted from images and a RUSBoost classifier derived computer fluorescence scores through cross-validation. Red fluorescence plaque levels increased during the 48-h accumulation. Plaque progression was identified by dentist assessment and computer analysis, presenting significant differences between visits at tooth and subject levels (p<0.05). Moderate correlations showed between clinical plaque and red fluorescence plaque (r=0.62 dentist, r=0.55 computer). The best agreement was observed when disclosing plaque threshold at level 2, for both dentist evaluation (sensitivity 71.1%, specificity 67.7%, accuracy 70.2%) and computer classification (sensitivity 68.4%, specificity 62.9%, accuracy 67.1%). Given the correlation with clinical diagnosis, red fluorescence imaging shows its potential for providing an objective and promising method for proper oral hygiene assessment.

Original languageEnglish
Article number096008
JournalJournal of Biomedical Optics
Issue number9
Early online date18 Sept 2017
Publication statusPublished - 2017


  • autofluorescence imaging
  • computer analysis
  • dental plaque
  • oral hygiene
  • RUSBoost classifier


Dive into the research topics of 'Red fluorescence imaging for dental plaque detection and quantification: pilot study'. Together they form a unique fingerprint.

Cite this