Date of Completion


Embargo Period



Patricia Diaz, Flavio Uribe

Field of Study

Dental Science


Master of Dental Science

Open Access

Open Access


A Clinical Trial to Investigate Digital Gingivitis Image Analysis Method and

Examiner-Based Grading in Assessing Experimental Gingivitis

Liliana Aranguren D.D.S

University of Connecticut, 2016

Objective: To compare a Digital Gingivitis Image Analysis (DGIA) method and examiner-based gingival grading in assessing experimental anterior gingivitis.

Methods: This was a single-center, examiner-blinded, parallel-group, randomized clinical trial. The study had 49 subjects and consisted of 2 phases: 2 weeks Oral Hygiene phase and 3 week induced gingivitis phase (days 0-21). During the hygiene phase, subjects received oral prophylaxis and used an electric toothbrush toothbrush. At day 0 of the Induced Gingivitis phase, 25 patients were randomly selected into the power brush group, which brushed twice daily. 22 Subjects, which abandoned all oral hygiene, were randomly selected into the “no hygiene” (control) group. Both control and experimental groups received an oral prophylaxis at day 21. Gingivitis was assessed via Loe-Sinless examinations (LSGI) and Digital Gingivitis Image Analysis at days -14, 0, 14 and 21.

Results: Groups were balanced (p > 0.4) with respect to demographics and gingivitis scores. At day 0, mean LSGI was 1.20 and mean number of bleeding sites was 40.49. Statistically significant (p < 0.001) differences in frequency of gingivitis were observed between the groups at days 14 and 21. At day 21, LSGI adjusted means were 1.16 and 1.56 for the power brush and the “no hygiene” group, respectively, corresponding to a 25% difference. Adjusted means for number of bleeding sites was 36.25 for power brush and 91.25 for the “no hygiene” group, corresponding to a 60% difference. Using the imaging system we noticed the marked changed in color from day 0 to day 21. We observed 2 units of color change (ΔE2000= 2.41) for non-brushing group and 1.4 units change on brushing group (ΔE2000 = 1.41). These changes were consistent with the clinical measurements.

Conclusion: The present study using the short-term, induced gingivitis model demonstrated Digital gingivitis image analysis positively associates with clinical parameters in an experimental gingivitis model

Major Advisor

Efthimia Ioannidou