Detection of Bacteria-Induced Early-Stage Dental Caries Using Three-Dimensional Mid-Infrared Thermophotonic Imaging

Abstract

Tooth decay, or dental caries, is a widespread and costly disease that is reversible when detected early in its formation. Current dental caries diagnostic methods including X-ray imaging and intraoral examination lack the sensitivity and specificity required to routinely detect caries early in its formation. Thermophotonic imaging presents itself as a highly sensitive and non-ionizing solution, making it suitable for the frequent monitoring of caries progression. Here, we utilized a treatment protocol to produce bacteria-induced caries lesions. The lesions were imaged using two related three-dimensional photothermal imaging modalities: truncated correlation photothermal coherence tomography (TC-PCT) and its enhanced modification eTC-PCT. In addition, micro-computed tomography (μ-CT) and visual inspection by a clinical dentist were used to validate and quantify the severities of the lesions. The observational findings demonstrate the high sensitivity and depth profiling capabilities of the thermophotonic modalities, showcasing their potential use as a non-ionizing clinical tool for the early detection of dental caries.

Keywords: (enhanced) truncated correlation photothermal coherence tomography; dental caries; linear iso phase (LIOP); thermophotonic; three-dimensional imaging.

Figure 1

Block schematic of the thermophotonic imaging system: the bacteria-induced caries is highlighted by the purple rectangle and is positioned normal to the MWIR camera.

Figure 2

Optical images of the (A) buccal, (B) distal, and (C) lingual surfaces of sample B8 after 8 days of cariogenic biofilm exposure. The lesions formed by the cariogenic biofilm exposure are outlined by the red rectangles.

Figure 3

µ-CT images of the B8 sample after 8 days of cariogenic biofilm exposure: Three lesions are visible, one on each of the buccal, distal, and lingual surfaces. The dashed black line corresponds to the cross-sectional plane shown in the transverse views. The colored rectangles on the transverse views outline the position of the lesions. The “Lesion View” column shows an enlarged view of the lesions. To increase the visibility of the lesions, an adaptive histogram equalization was used [30] to improve the image contrast and visualize the edges of the lesion more definitively.

Figure 4

The first slices of the eTC-PCT amplitude images of sample B8: The three images show the (A) buccal, (B) distal, and (C) lingual surfaces of the sample. The red rectangle (10 × 5 pixels) highlights the region of the caries lesion used to determine the carious signal response of the sample, and the dashed blue rectangle (10 × 5 pixels) highlights a nearby sound region used as the reference for the sound enamel signal.

Figure 5

Mean amplitude and phase responses of the caries lesions and sound enamel regions highlighted in Figure 4 for (A) TC-PCT amplitude, (B) eTC-PCT amplitude, (C) TC-PCT phase, (D) eTC-PCT phase, and (E) TC-PCT LIOP.

Figure 6

eTC-PCT amplitude reconstructions with partial transparency of the (A) buccal, (B) distal, and (C) lingual surfaces using a 0.2–0.6 Hz, 12 s chirp with 40 ms pulse width and 20 ms slice width for reconstructions. The dashed black lines indicate the regions where the cross-sectional cuts (displayed underneath the whole volume) were made. The reconstructions were generated by linearly rescaling each tomographic slice between (0,1) and stacking the layers. Volumes are displayed using software available online [31].

Figure 7

TC-PCT LIOP reconstructions with partial transparency of the (A) buccal, (B) distal, and (C) lingual surfaces using a 0.2–0.6 Hz, 12 s chirp with 40 ms pulse width and 20 ms slice width. All reconstructions were filtered using the histogram equalization filter [13] to improve the caries visibility. The dashed black lines indicate the regions of the cross-sectional cuts (displayed underneath the whole volume). Volumes are displayed using software available online [31].

Figure 8

Select tomographic phase slices: (A) eTC-PCT phase, (B) TC-PCT phase, and (C) TC-PCT LIOP of the B8 sample lingual surface after 8 days of bacteria-induced demineralization. Their corresponding phase values were normalized between 1.0 and 0 by subtracting the minimum phase value and then dividing by the maximum phase value within each slice. All slices were calculated using the same data from a 0.2–0.6 Hz, 12 s chirp with 63 W peak laser power. Contrast of the slices has been enhanced using the slice-by-slice global histogram equalization algorithm described in software available online [13].

Figure 9

Optical images of samples M2, M4, M6, and M8 after cariogenic biofilm exposure. The red rectangles show the regions of exposure on the teeth. Only sample M8 (8 days of cariogenic biofilm exposure) developed visible caries.

Figure 10

First slice of the eTC-PCT amplitude for samples (A) M2, 2 days of treatment, (B) M4, 4 days of treatment, (C) M6, 6 days of treatment, and (D) M8, 8 days of treatment. All images were acquired using a 0.2–0.6 Hz, 12 s chirp with 40 ms pulse widths, 63 W peak laser power and 20 ms slice widths for reconstructions. The red rectangle highlights the treatment region where the samples were exposed to the cariogenic biofilm. The dashed blue rectangle highlights the nearby sound region that was not exposed to the cariogenic biofilm. This region was used to determine a baseline sound enamel signal.

Figure 11

Mean amplitude and phase responses of the caries lesions and sound enamel regions highlighted in Figure 10 for (A) TC-PCT amplitude, (B) eTC-PCT amplitude, (C) TC-PCT phase, (D) eTC-PCT phase, and (E) TC-PCT LIOP.

Figure 12

µ-CT results of sample M8 (8 days of demineralization): (A) Volume reconstruction with no visible lesion. The dashed blue oval indicates the region where the lesion is visible with the naked eye, the dashed red line indicates the plane where the transverse view B is taken. (B) Transverse view of the sample with the red box indicating where the lesion should be visible. (C) Enlarged view of the red box seen in B, with no lesion visible.

Figure 13

Amplitude reconstructions of the M8 sample (8 days of cariogenic biofilm exposure) using the (A) eTC-PCT and (B) TC-PCT algorithms with 20 ms slice widths. The amplitudes were normalized between 1.0 and 0 by subtracting the minimum amplitude value and then dividing by the maximum amplitude value within each slice. Volumes are displayed using the method of [31].