Artificial intelligence for early diagnosis and risk prediction of periodontal-systemic interactions: Clinical utility and future directions

Table 1 A Comparison of the effectiveness of artificial intelligence models in diagnosing periodontal-systemic interactions across various studies

Study	Sample size	Diagnostic accuracy (%)	Sensitivity (%)	Specificity (%)	AI model type
Study 1	120	85	80	90	Machine learning
Study 2	350	88	85	88	Deep learning
Study 3	800	90	87	92	Deep learning
Study 4	1500	92	90	94	Machine learning
Study 5	1000	87	83	89	Machine learning

AI: Artificial intelligence.

Table 2 Subgroup analysis of cone beam computed tomography, magnetic resonance imaging, and artificial intelligence tools in diagnostic error reduction, soft-tissue diagnosis, and diagnostic time efficiency

Modality	Reduction in diagnostic errors (%)	Improvement in soft-tissue diagnostics (%)	Reduction in diagnostic time (%)
CBCT	35	N/A	N/A
MRI	N/A	25	N/A
AI tools	N/A	N/A	40

CBCT: Cone beam computed tomography; MRI: Magnetic resonance imaging; AI: Artificial intelligence; N/A: Not applicable.