The use of artificial intelligence in radiological diagnosis and detection of dental caries: a systematic review

Abstract

Dental caries is a very common condition, which can lead to serious complications, including tooth loss and infection of the whole human body. Dentists in their daily practice, apart from visual-tactile examination, use radiological methods, such as periapical radiographs and bitewings. Artificial intelligence (AI) is a tool that can be used in diagnosing and detecting cavities. It can help to avoid more invasive treatment and further consequences. The goal of this systematic review was to present the use of artificial intelligence in radiological dental caries diagnostics. In total, twelve studies meeting inclusion criteria were analyzed, and image databases varied from 93 to 3,868 radiographs, with average value of 1,091.17 radiographs. Most of the included studies employed bitewings and periapical images, and authors used different methods and AI algorithms. Accuracy was performed in nine researches. The highest accuracy was 99%, the lowest 73.3%. Also, nine researches provided information on number of observers, which varied from 1 to 25. Comparing all the studies, it was difficult to draw out a conclusion. Artificial intelligence in radiological images may assist dentists and radiologist to perform better and faster examination, and it may be a used in routine dental care. However, more researches are needed in the field of dentistry and radiology.

Ähnliche Arbeiten