Introduction: The CNN–Fuzzy approach is used in this article to create a computer-aided clinical decision-making system. According to the literature, dentists are inconsistent when it comes to diagnosing abscesses, fractures, impacted teeth, reversible and irreversible pulpitis. As a result, the objective of this research is to guide and assist dentists in more accurately diagnosing dental diseases. Methods: To address inaccurate and ambiguous dental radiograph values, as well as disease signs and symptoms, a robust algorithm based on CNN–Fuzzy logic has been developed. To initiate, the probability of diseases was calculated for each category using an independently designed CNN approach, which was then applied to a fuzzy knowledge base and the Mamdani inference, which contains 947 rules to diagnose diseases and make recommendations to the dentist. Results and Discussion: The CNN–Fuzzy approach's findings are compared to dentists' recommendations. With the help of five professional dentists, the accuracy, macro-averaged precision, recall, F1 score, and kappa value are calculated from 250 randomly generated sample cases. The CNN–Fuzzy approach has a 92.8% accuracy, which is 6.4% higher than specialist prediction. The proposed method yields result that are consistent with the dentists' diagnoses. Conclusion: The proposed computer-aided decision-making system for dental diseases boosts dentists' confidence in diagnosing abscesses, fractures, impacted teeth, reversible and irreversible pulpitis, and reduces false diagnoses caused by ambiguous values of dental radiograph, signs, and symptoms.