Spectral characteristics of caries-related autofluorescence spectra and their use for diagnosis of caries stage

Deep learning and sub-band fluorescence imaging-based method for caries and calculus diagnosis embeddable on different smartphones

Popularizing community and home early caries screening is essential for caries prevention and treatment. However, a high-precision, low-cost, and portable automated screening tool is currently lacking. This study constructed an automated diagnosis model for dental caries and calculus using fluorescence sub-band imaging combined with deep learning. The proposed method is divided into two stages: the first stage collects imaging information of dental caries in different fluorescence spectral bands and obtains six-channel fluorescence images. The second stage employs a 2-D-3-D hybrid convolutional neural network combined with the attention mechanism for classification and diagnosis. The experiments demonstrate that the method has competitive performance compared to existing methods. In addition, the feasibility of transferring this approach to different smartphones is discussed. This highly accurate, low-cost, portable method has potential applications in community and at-home caries detection.

Machine learning-based automatic identification and diagnosis of dental caries and calculus using hyperspectral fluorescence imaging

PURPOSE

Precise diagnosis and identification of early dental caries facilitates timely intervention and reverses the progression of the disease. Developing an objective, accurate and rapid caries and calculus automatic identification method advances clinical application and facilitates the promotion and screening of oral health in the community and family.

METHODS

In this study, based on 122 dental surfaces labeled by professional dentists, hyperspectral fluorescence imaging combined with machine learning algorithms was employed to construct a model for simultaneously diagnosing dental caries and calculus. Model trained by fusion features based on spectra, textures, and colors with the integrated learning algorithm has better performance and stronger generalization capabilities.

RESULTS

The experimental results showed that the diagnostic model's accuracy, sensitivity, and specificity for identifying four different caries stages and calculus were 98.6%, 98.4%, and 99.6%, respectively.

CONCLUSIONS

The proposed method can evaluate the whole tooth surface at the pixel level and provides discrimination enhancement and a quantitative parameter, which is expected to be a new approach for early caries diagnosis.

Fluorescence spectrometry based chromaticity mapping, characterization, and quantitative assessment of dental caries

PURPOSE

Dental caries detection, especially the accurate detection of early caries, facilitates prompt interventions. It is reasonably common to use fluorescence imaging for classification and evaluation of caries, but lacks a quantitative, precise and easy-to-use characterization for practical applications. In this study a quantitative approach for caries stage detection by correlating caries spectral and chromatic features was examined.

METHODS

A 405 nm LED light source was used as the excitation source. A hyperspectral imaging camera is employed to collect 336 spectral data of different caries stages. Four critical intervals for different stages of caries were extracted by fluorescence spectral features. The mapping relationship between caries spectral and chromatic features was established by Fast Formula Fitting (FFF) and Neural Network Fitting (NNF) methods.

RESULTS

The 470-780 nm spectral power distribution was proved to be the best matching color waveband guiding the selection of filters in future instrument development. The correlation coefficients for the two fitting methods were 0.990 and 0.999, respectively. Both methods achieved caries stage prediction at the pixel level with high accuracy using color information. The visualization region in the chromaticity diagram was created.

CONCLUSIONS

This quantitative method enables accurate prediction of caries on the entire tooth surface and facilitates the development of portable and low-cost caries detection instruments.

Analysis of enamel/restoration interface submitted cariogenic challenge and fluoride release

The treatment of high-risk patients still is a challenge. The understanding and development non-invasive, non-destructive, and non-ionizing techniques, can help to guide the treatment and the diagnosis of primary and recurrent caries. The present study evaluated the behavior of enamel/restoration interface after a cariogenic challenge by Fourier domain optical coherence tomography (FD-OCT), scanning electron microscopy (SEM) examination, and the fluoride release of the different restorative materials. Cavities (1.5 × 0.5 mm) were performed in enamel surface and divided into groups (n = 8): glass ionomer cement (GIC), resin-modified glass ionomer cement (RMGIC), and resin composite (RC). The samples were submitted to pH-cycling, and the solutions analyzed for cumulative fluoride by ion-analyzer. The morphology was analyzed by SEM through replicas. The optical attenuation coefficient (OAC) was calculated through exponential decay from the images generated by FD-OCT. Data were analyzed considering α = 0.05. OAC values increased for all groups after pH-cycling indicating demineralization (p < .05). Considering the remineralizing solution, RMGIC presented higher fluoride release rate, followed by GIC, while RC did not release any fluoride. Yet for the demineralizing solution, RMGIC and GIC released similar fluoride rates, overcoming RC (p < .05). Micrographs revealed no changes on the restorations margins, although enamel detachment was observed for RC and GIC after pH-cycling.

Application of optical and spectroscopic technologies for the characterization of carious lesions in vitro

The detection of the beginning demineralization process of dental hard tissue remains a challenging task in dentistry. As an alternative to bitewing radiographs, optical and spectroscopic technologies showed promising results for caries diagnosis. The aim of the present work is to give an overview of optical and spectroscopic properties of healthy and carious human teeth in vitro by means of Raman spectroscopy (RS), optical coherence tomography (OCT) and hyperspectral imaging (HSI). OCT was able to represent microstructural changes below the enamel surface and revealed increased scattering for white spot lesions and a white scattering trail for deeper lesions. HSI showed similar absorbance characteristics for healthy and demineralized enamel over the entire spectrum and a characteristic absorbance peak at 550 nm for discolored lesions. Already at early carious stages (white spot), we found a distinct loss of hydroxylapatite-related intensity at 959 cm−1 in demineralized regions with RS. Healthy and demineralized tooth surfaces can be distinguished at different signal levels by means of RS, OCT and HSI. The presented modalities provide additional information to the current clinical diagnosis of caries such as microstructural changes, quantification of the demineralization and imaging of caries-related chemical changes.

Fluorescence change of Fusobacterium nucleatum due to Porphyromonas gingivalis

The aim of this study was to measure changes in the fluorescence of Fusobacterium nucleatum interacting with Porphyromonas gingivalis for excitation with blue light at 405-nm. P. gingivalis was mono- and co-cultivated in close proximity with F. nucleatum. The fluorescence of the bacterial colonies was photographed using a QLF-D (Quantitative Light-induced Fluorescence-Digital) Biluminator camera system with a 405 nm light source and a specific filter. The red, green and blue intensities of fluorescence images were analyzed using the image analysis software. A fluorescence spectrometer was used to detect porphyrin synthesized by each bacterium. F. nucleatum, which emitted green fluorescence in single cultures, showed intense red fluorescence when it was grown in close proximity with P. gingivalis. F. nucleatum co-cultivated with P. gingivalis showed the same pattern of fluorescence peaks as for protoporphyrin IX in the red part of the spectrum. We conclude that the green fluorescence of F. nucleatum can change to red fluorescence in the presence of adjacent co-cultured with P. gingivalis, indicating that the fluorescence character of each bacterium might depend on the presence of other bacteria.

Difference assessment of composite resins and sound tooth applicable in the resin-imbedded tooth for resin repair using fluorescence, microhardness, DIAGNOdent, and X-ray image

OBJECTIVE

Visual differentiation of resin and tooth in a tooth cavity is not simple due to their highly similar shade. The purpose of the present study was to find any noninvasive method which can effectively differentiate resin from sound tooth in a resin-imbedded tooth for resin repair.

MATERIALS AND METHODS

For the study, various resin products were imbedded into the cavity of sound tooth. By applying laser of different wavelengths, autofluorescence (AF) of sound tooth and resin products were obtained. Microhardness, X-ray radiograph, and DIAGNOdent were tested for each tooth, resin product, and resin-imbedded tooth.

RESULTS

For the AF spectra obtained using the 405-nm wavelength, sound tooth has emission peak at 440-470 nm and near 490 nm. Sound tooth has several times higher microhardness than resin products regardless of position in tooth subsurface. Due to the difference of radiopaque fillers' composition and concentration, resin products have different brightness in the X-ray radiograph. DIAGNOdent readings for tooth and resin products were inconsistently different, and the difference of obtained values was slightly not to be applicable for the differentiation.

CONCLUSION

Among the tested methods, with noninvasive treatment, AF spectrum by the 405-nm wavelength showed the apparent difference between resin and tooth.

CLINICAL SIGNIFICANCE

For the resin repair in a resin-imbedded tooth cavity, AF spectrum produced by 405-nm wavelength could be a useful method in tracing the resin-tooth boundary if combined with conventional X-ray radiography.

Diagnosis and staging of caries using spectral factors derived from the blue laser-induced autofluorescence spectrum

OBJECTIVE

The aim of this study was to identify the factors derived from the 405nm laser-induced autofluorescence (AF) spectra that could be used to diagnose and stage caries.

MATERIALS AND METHODS

Teeth (20 teeth per stage) were classified as sound, stage II, III, and IV based on a visual and tactile inspection. The specimens were re-examined and reclassified based on micro-CT analysis. From the teeth, the AF was obtained using a 405nm laser. Three spectral factors (spectral slope at 550-600nm, area under the curve at 500-590nm, and two-peak ratio between 625 and 667nm) were derived from the AF spectra. Using these factors, the diagnosis and staging of caries were tested, and the results were compared with those of DIAGNOdent.

RESULTS

After micro-CT analysis, only 13, 11, and 13 teeth were reclassified as stages II, III, and IV, respectively. The reclassified groups showed less data overlap between the stages, and the spectral slope was 40.1-74.6, 27.5-39.6, 11.1-27.4, and 1.0-9.7 for sound, stage II, III, and IV, respectively. The differentiation of stages III and IV using DIAGNOdent appeared to be difficult due to the considerable data overlap.

CONCLUSION

Among the factors tested, the spectral slope at 550-600nm showed the best match with the caries specimens, in which their stage had been identified precisely.

CLINICAL SIGNIFICANCE

The 405nm laser-induced AF spectra can be applied to the diagnosis and staging of caries alone or in conjunction with conventional methods, such as visual, tactile, and X-ray inspection.