shilpansingh / cavity-detection-using-ai

Here is a simple implementation for detection of tooth cavity using Ai

Cavity-Detection-Using-Ai

Here is a simple implementation for detection of tooth cavity using Ai

Abstract:

The system is developed in such a way that it can tell the presence and the intensity of the cavity or caries. We wish to train the model using CNN which is a deep learning approach. Here we train the models with a dataset of over 500 images, for improved accuracy. The developed system would make it easy and very convenient to detect the cavities in any stage and also determine their severity.

Introduciton

History has shown that untreated caries places a significant biological, social, and financial burden on healthcare systems in general and on individuals. Many individuals with untreated caries wait until it is too late to seek treatment because it is not life-threatening, which can result in substantial acute and chronic conditions. Problems that require expensive therapy. Consequently, an incredibly desirable fundamental issue for dental caries prevention techniques, such as fissure sealants, fluoride remineralization therapies, or other demineralizing products, is early detection of dental caries. A suitable and accurate approach of caries detection is urgently needed to improve clinical decision making in dental clinics as well as in oral epidemiology or caries research. Convolutional neural networks (CNNs), which have grown rapidly in popularity in the field of deep machine learning, are a particularly effective way for categorizing and analyzing images. Also, a recent study noted that these innovative algorithms can provide a financially viable answer for disease identification. As a result, it has gained popularity in medical analysis. For instance, it was used in the identification of malignant tumor cells and the classification of diabetic retinopathy. CNNs have become known for their excellent recognition rates, precision, and effectiveness thanks to these applications. Dental image analysis has been criticized for its low reliability and validity, and artificial intelligence (AI), more specifically deep learning utilizing convolutional neural networks (CNNs), has been proposed as a potential solution. With a collection of weights that have been learned from data, CNNs enable the mapping of an input to an output.

How to run this project?

1. Download the entire repo.
2. Create a virtual environment for running python in your desired folder.
3. Host the application usign streamlit as a local host.