This study aims to develop an AI program that can classify oral findings into Normal/variation of normal or an oral disease by clinical photos analysis, aiding in lowering the percentages of false positive and false negative diagnosis of oral diseases.

Early diagnosis of oral lesions, particularly oral cancer, is crucial for enhancing prognosis, facilitating early intervention and care with the intention of lowering disease-related mortality.

Since conventional oral examination (COE) is the most used method in identifying oral lesions, the average dental practitioner's experience is a decisive factor in early diagnosis.

Visual examination lacks specificity and sensitivity since its highly subjective. Unfortunately, Studies show that the majority of dentists lack expertise in early detection of the disease, resulting in false negative diagnosis of oral lesions.

General practitioners are found to either delay the referral of a suspected oral lesion to an Oral Medicine specialist, or referring numerous false positive cases, unnecessarily pushing the patients into a state of anxiousness and cancer phobia. False positive referrals overburden the specialists, which will eventually cause delayed diagnosis of true positive cases due to the oversaturation with false positive ones.

diagnostic research scope shifts towards noninvasive, easy chair side methods with higher accuracy for early detection of oral lesions. Recent approaches towards using machine based programs indicate that this machine-learning method may be useful in the detection and diagnosis of oral cancer.

Inclusion Criteria:

• Patients above 18 years old
• Candidates with normal oral cavity findings
• Candidates with variations of oral cavity findings
• Candidates with different oral lesions

Exclusion Criteria:

• Patients less than 18 years old

patients should be above 18 years old, with no maximum age limit. Normal oral cavity findings, variations of oral anatomical landmarks, patients with oral lesions are all included in the study.