Convolution Neural Networks and Targeted Fluorescent Nanoparticles to Detect and ICDAS Score Caries.

Previous work has shown targeted fluorescent starch nanoparticles (TFSNs) can label the subsurface of carious lesions and assist dental professionals in the diagnostic process. In this study, we aimed to evaluate the potential of using artificial intelligence (AI) to detect and score carious lesions using ICDAS in combination with fluorescent imaging following application of TFSNs on teeth with a range of lesion severities, using ICDAS-labeled images as the reference standard. A total of 130 extracted human teeth with ICDAS scores from 0 to 6 were selected by a calibrated cariologist. Then, the same surface was imaged with a stereomicroscope under white light illumination, without visible fluorescence, and blue light illumination with an orange filter following application of the TFSNs. Both sets of images were labeled by another blinded ICDAS-calibrated cariologist to demarcate lesion position and severity. Convolutional neural networks, state-of-the-art models in imaging AI, were trained to determine the presence, location, ICDAS score (severity), and lesion surface porosity (as an indicator of activity) of carious lesions, and tested by 30 k-fold validation for white light, blue light, and the combined image sets. The best models showed high performance for the detection of carious lesions (sensitivity 80.26%, PPV 76.36%), potential for determining the severity via ICDAS scoring (accuracy 72%, SD 5.67%), and the detection of surface porosity as an indicator of the activity of the lesions (accuracy 90%, SD 7.00%). More broadly, the combination of targeted biopolymer nanoparticles with imaging AI is a promising combination of novel technologies that could be applied to many other applications.

Early occlusal caries detection using targeted fluorescent starch nanoparticles.

OBJECTIVES: We have previously shown fluorescent cationic starch nanoparticles (FCSNs) penetrate enamel surface porosity of active carious lesions, potentially aiding their detection. Here, we evaluate the in vitro diagnostic accuracy of FCSNs in detecting occlusal caries compared to histologic reference standard. METHODS: 100 extracted human teeth were selected with sound (50), or either non-cavitated (25) or cavitated (25) lesions. A region of interest (ROI) on the occlusal surface was assessed for fluorescence by two independent examiners, after immersion in FCSN solution, water rinse, and illumination by dental curing lamp viewed through orange UV-filter glasses. ROIs were sectioned and evaluated by histology (Downer Criteria) as a gold standard for caries presence. Cohen's Kappa was determined for inter- and intra-examiner agreement, and sensitivity, specificity, and area under the curve of Receiver Operator Curves (ROCAUC) were calculated. The analysis was repeated for the subset of "early" lesions, defined as being limited to enamel. RESULTS: FCSN use resulted in substantial inter-user (k=0.74±0.07), and high intra-user agreement (k=0.80±0.06; 0.94±0.03, by examiner). Sensitivity, specificity and ROCAUC for FCSNs were 88.9%; 94.6%; 0.92±0.06 for all, and 76.9%, 94.6%, and 0.86±0.10 for early lesions. In post hoc analysis, sensitivity seemed to be greater with the FCSN than the expert visual exam, particularly for early lesions. CONCLUSIONS/CLINICAL SIGNIFICANCE: FCSNs are a reproducible and accurate novel technology for occlusal caries detection, with high sensitivity and specificity compared to histology. Future clinical validation is necessary. FCSNs can improve early caries detection and shift treatment towards non-invasive approaches, improving oral health.