Common dental diseases like dental cavities can cause discomfort, pain and ultimately tooth loss. These detrimental effects can be avoided with early dental disease detection and treatment. The hardest material in the human body is the tooth. Current techniques for identifying dental issues are characterized by their high degree of user intervention, intricacy of the experience operation, and low efficiency. Maintaining a patient’s dental health and averting further issues depends on the early detection of tooth-related disorders. Many patients miss out on timely treatment because dentists are not always aware of tooth-related conditions that can be hard to detect visually. Earlier methods of detecting dental diseases were labour-intensive and necessitated a dentist’s examination and assessment of the condition. There are many difficulties in diagnosing and identifying dental issues. In the past, diagnosing dental diseases was a laborious manual procedure that required dentists to carefully inspect and assess the situation. One revolutionary method to support medical imaging diagnosis is the incorporation of Artificial Intelligence (AI). Analysing dental images is a crucial part of the diagnosis process in routine clinical practice. Throughout the diagnostic procedure, the dentist has to analyse many tooth-related issues, such as tooth counts and associated disorders. Nowadays, a significant portion of the population suffers from dental issues, which are a widespread global health concern. Dental diseases are effectively treated to prevent further problems, so an early and accurate diagnosis is crucial. In a range of health imaging claims, deep learning algorithms have recently shown incredible efficacy. Human dental disease is a serious issue, and deep learning is being applied in dentistry with increasing frequency. This work aims to explore the potential of deep learning for the detection of dental diseases using optical images. This research proposed an innovative AI model for diagnosing dental disease from optical imaging. The required optical images are collected from the online resources. The collected optical images are directly given in to VGG16 Conditional Autoencoder with Res-LSTM (VGG16-CAE-RLSTM). The incorporation of different deep learning models offers an easy and efficient way to diagnose the dental disease in the human and it is helpful for the dentist to adopt the treatment to solve different kinds of dental problems in the human. As a part of the research, the proposed model is evaluated with the prior models to showcase the efficiency in dental disease diagnosis.