Project Home
PDF access


Aly Yasser, Sandy Elias George, Anas El Shazly, Yasmina Basim,Dr.Ammar Mohamed, Eng. Nour El-Huda

Publishing Date

May 14, 2022


Periodontal disease (PD) is characterised by gingival inflammation, connective tissue loss, or alveolar bone degradation. It’s the world’s sixth most common inflammatory disease. Periodontal disease is the leading cause of tooth loss, and it has been linked to heart disease, diabetes, and rheumatoid arthritis. Clinical examination and radiograph analysis are two of the most popular approaches used nowadays to diagnose Periodontal disease. Due to the constraints of the standard diagnostic paradigm, inaccuracy due to human error, and conflicting judgements by various examiners. As a result, more consistent method of identifying Periodontal Disease is required. Because of advancements in machine learning research, automated medical support systems are in high demand for screening at-risk individuals for periodontitis, and early detection may postpone the beginning of tooth loss, particularly in small communities and health care settings settings with a shortage of dental specialists. We present a periodontal disease detection as well as severity prediction approach based on Deep Learning.

1.1 Background

Periodontal disease is an inflammatory pathological disease that affects the gums and bone support surrounding the teeth, causing tooth loss. The intensity is determined by the patient’s actions (such as smoking) as well as medical issues (such as diabetes). Periodontal problems are divided into two categories: gingivitis and periodontitis. [kononen2019periodontitis] This kind of periodontitis is characterised by the loss of periodontal ligaments and the resorption of the surrounding alveolar bone as a result of severe inflammatory processes. The loss of alveolar bone weakens the teeth’s supporting structures, making the patient more vulnerable to tooth movement and loss. The dentist examines the patient’s medical history for any factors that may be contributing to the symptoms, such as smoking or using medications that cause dry mouth. The dentist examines the teeth for plaque and tartar accumulation, as well as any signs of bleeding. detection takes place by Placing a dental probe next to the tooth beneath the gum line, often at many locations across the mouth, to measure the pocket depth of the groove between the gums and teeth. The pocket depth in a healthy mouth is normally between 1 and 3 millimeters (mm). Periodontitis may be indicated by pockets deeper than 4 mm. Cleaning pockets with a depth more than 5 mm is impossible. The dentist will take dental X-rays to check for bone loss in places where deeper pockets have been discovered. As a result, early detection is critical for preventing periodontitis, especially in community and health-care settings where dentists are hard to come by. The goal of this study is to develop an interpretable computer-aided diagnosis approach for periodontitis based on panoramic radiographs. Dentists frequently need to serve a large number of patients and examine a large number of panoramic radiographs on a regular basis in order to avoid inaccuracy due to human error.

1.2 Motivation


• Patients are constantly presented with medical apparatus and invasive operations (e.g., needle injections and extractions) during dental treatment, which can easily lead to discomfort and the development of dread and phobia. Indeed, demographic surveys reveal that more than 5% of individuals completely avoid going to dental care, while 20 to 30% dread going to the dentist so much that they only visit , when they are in a lot of pain or discomfort[de1998blood].In addition to bleeding on probing (BOP) is a common criteria for determining gingival inflammation.

• Periodontal disease has been identified as the leading cause of tooth loss in adults. The continuous progression of PD(Periodontal disease )results in the destruction of all periodontal supporting tissues, including the alveolar.

• A high level of clinical examination competence is required to detect and diagnose periodontal disease. Periodontal disease is discovered utilising a dental probe positioned adjacent to the tooth to measure the depth of the groove between the gums and teeth beneath the gum line in a number of locations across the mouth, which may be painful for the patient. Most specialists believe that radiographs may be used to supplement clinical examinations in order to make a diagnosis and guide a patient’s treatment approach although to prevent human errors. Meanwhile the radiographs does not show the severity degree of periodontitis.


Periodontitis is a common and irreversible chronic inflammatory illness that affects 20–50% of the worldwide population in both industrialized and developing countries. Periodontal disease is caused by bacteria found in the periodontal pocket. Local communities with a limited number of healthcare providers require an automated computer-aided diagnosis approach for periodontal disease, as it is physically exhausting for dentists to utilize dental probing or examine radiographs individually. The periodontal probe is a little spherical instrument with a blunt end measured in millimeters, which the dentist carefully inserts between the tooth and gingiva until a minor resistance is felt. Aside from redness, bleeding in response to moderate proving is another sign of gingivitis. while also reducing human error. The proposed system will detect whether the teeth are healthy or periodontal, as well as the severity degree.

1.3 Problem Statement

• For both the patient and the dentist, measuring the groove between the gums’ pocket depth and teeth, using a dental probe at many locations can be exhausting.

• The inability to estimate the severity of periodontitis.

• Detecting periodontitis in early stages in order to prevent teeth loss.