Background : Identification of dental implants using deep learning
Background
Dental implants were developed in the 1980s, and they are now used for patients with missing teeth globally. Their effect on dental treatment is great, and various improvements in patients’ quality of life have been reported. Implant treatment is no longer unusual for either patients or dentists. However, because more than 30 years have passed since implants were introduced into clinical practice, various implant problems have been also reported, such as complications in the superstructures or implants and peri-implantitis. Additional prosthodontic, periodontic, or surgical treatments are needed to solve these problems. When performing these treatments, various information is needed about the intra-oral implant, such as the manufacturer, system, length, and width of implant, method of fixation, and type of abutment. If the implant patient was previously treated at the same family clinic, this information is easy to obtain from the patient’s medical record. However, if the treatment was performed at another clinic and the patient cannot contact the treatment provider, this information may be difficult or impossible to obtain. Recently, some patients with implant troubles have visited other clinics for various reasons, such as relocation or the closure of family clinics. In such cases, dentists must identify the patient’s implant information from the limited data obtained from oral photographs, radiographs, a study model, and so on. In particular, the type of implant system must be identified in order to conduct additional treatments. Dentists with sufficient knowledge about and experience of implant treatments can identify implant systems and perform treatment, but those without the knowledge and experience cannot identify the system and treat the patient. Therefore, there is a need for a system for identifying the implant system of a patient from limited data that does not depend on the dentist’s knowledge and experience.
Artificial intelligence (AI) technology has been applied in various fields, and its presence is already essential in many of them. In AI technology, there are several methods that are used in accordance with the task. In medicine, AI has already been used for robotics, medical diagnosis, statistics, and human biology-up. A deep learning method, one of the AI technologies, is adequate for prediction, object detection, classification, and other similar tasks. In dentistry, the diagnosis of dental diseases using oral or X-ray images, prediction of treatments, classification, statistics from research data, and other topics have been addressed using a deep learning method. Specifically, studies on the diagnosis of diseases using a deep learning have increased, and deep learning-based object detection algorithms for images are usually used for this task. The ability of diagnostic systems using deep learning is already comparable or superior to that of humans, and these systems will help prevent dentists from missing problems or making errors. If this system also can be applied for identifying implant systems using dental X-ray images, it will help both dentists and patients solve implant problems.
The purpose of this study is to develop an automated system for identifying implant systems using a deep learning-based object detection method. The hypothesis of this study was that this system could detect and identify the implant.
Serial posts:
- Identification of dental implants using deep learning — pilot study
- Background : Identification of dental implants using deep learning
- Method : Identification of dental implants using deep learning
- Results & Discussion : Identification of dental implants using deep learning (1)
- Results & Discussion : Identification of dental implants using deep learning (2)
- Results & Discussion : Identification of dental implants using deep learning (3)
- Conclusion : Identification of dental implants using deep learning
- Figure 1: Sample image for calculating IoU (MK III implant)
- Figure 2. Total number of objects of each implant systemin all images.
- Figure 3. The total number of implant systems detected correctly (TPs)
- Figure 4. Ratio of implant systems detected correctly to all detected systems
- Figure 5. Average precision (AP) of each implant system in all images
- Figure 6. Sample images of misdetected implants