Advancements in artificial intelligence and embedded systems have paved the way to cost-effective diagnostic tools in dentistry. In this work, we developed an Arduino-controlled rotational imaging device called DentalLens, combined with a convolutional neural network (CNN)-based algorithm to diagnose oral diseases. Introducing Dental Lens, a web app providing free initial dental consultations in the Philippines through photo evaluations. The prototype juxtaposes an Arduino-driven rotation imaging rig for image collection along with an AI (a hybrid ensemble model combining EfficientNetV2 and LightGBM)-based diagnostic method. This algorithm classifies images in six common oral diseases (Dental Calculus, Dental Caries, Gingivitis, Hypodontia, Mouth Ulcer, and Tooth Discoloration), returning diagnostic reports including the confidence scores. Developed for early alert and awareness, not professional diagnosis, Dental Lens is an example of how low-cost technology combined with AI can raise oral health literacy and enable faster intervention, particularly among underserved communities. An experimental study on an intraoral dataset realized by images showed its promising value, with the overall accuracy of CNN at 92.48% accuracy, with a mean confidence of 97.02% and a misclassification rate of only 7.52%. The result reveals that the DentalLens hybrid model could provide adequate diagnostic support.
Guirigay et al. (Wed,) studied this question.