In the era of rapid digitalization contemporary dentistry is developing as one of the most dynamic fields of medical practice, actively integrating digital technologies and interdisciplinary approaches.<\/p>\n
Artificial intelligence<\/strong> and modern digital platforms are becoming key components of the transformation of imaging-diagnostic processes in temporomandibular joint pathology.<\/p>\n The systematic review included five studies in which CBCT<\/strong>, MRI<\/strong> and panoramic radiography<\/strong> were used, demonstrating diagnostic accuracy ranging from moderate to high depending on the target task and study methodology.<\/p>\n The authors note that the current literature reflects a profound transformation of approaches to TMJ imaging \u2014 from manual interpretation of images to hybrid workflows, where AI<\/strong> provides increased reproducibility, detection of early osteoarthritic and morphological changes, as well as support for clinical decision-making; at the same time, standardization of model training protocols and external validation on transnational patient samples are critically important.<\/p>\n Within the review the role of multimodal approaches is traced, where CBCT and MRI data complement clinical examination information and anamestic data for comprehensive assessment of the TMJ.<\/p>\n CBCT<\/strong> combined with machine learning algorithms was most often used to navigate the joint anatomy and detect structural changes, such as erosions of articular surfaces, subchondral sclerosis, osteophytes and deformities of the condylar region; in clinical practice this opens the possibility of earlier identification of patients with progressive osteoarthritis and planning corrective interventions.<\/p>\n MRI<\/strong> remains the gold standard for assessment of the soft tissue components of the TMJ \u2014 the disc, the retrodiscal space, synovial fluid; AI-based MRI methods are capable of increasing sensitivity to internal derangements, however limitations are the quality and standardization of scanning protocols. Panoramic radiography<\/strong> provides limited useful information for structural analysis of the TMJ, AI applications on panoramic images are so far less informative for early diagnosis of intra-articular changes.<\/p>\n Key barriers to the implementation of AI in clinical practice remain small sample sizes, lack of external validation and variability of imaging protocols; to overcome these limitations multicenter studies, synchronization of clinico-radiological protocols and inclusion of models within existing diagnostic criteria are necessary.<\/p>\nArtificial intelligence as a key tool<\/h2>\n
Diagnostic ecosystem: structure and components<\/h2>\n
CBCT: identification of degenerative and morphological changes<\/h3>\n
MRI and panoramic radiography: functions and limitations<\/h3>\n
Validation, standardization and clinical integration<\/h2>\n