Artificial intelligence in detecting temporomandibular joint osteoarthritis on orthopantomogram

Abstract

Orthopantomogram (OPG) is important for primary diagnosis of temporomandibular joint osteoarthritis (TMJOA), because of cost and the radiation associated with computed tomograms (CT). The aims of this study were to develop an artificial intelligence (AI) model and compare its TMJOA diagnostic performance from OPGs with that of an oromaxillofacial radiology (OMFR) expert. An AI model was developed using Karas’ ResNet model and trained to classify images into three categories: normal, indeterminate OA, and OA. This study included 1189 OPG images confirmed by cone-beam CT and evaluated the results by model (accuracy, precision, recall, and F1 score) and diagnostic performance (accuracy, sensitivity, and specificity). The model performance was unsatisfying when AI was developed with 3 categories. After the indeterminate OA images were reclassified as normal, OA, or omission, the AI diagnosed TMJOA in a similar manner to an expert and was in most accord with CBCT when the indeterminate OA category was omitted (accuracy: 0.78, sensitivity: 0.73, and specificity: 0.82). Our deep learning model showed a sensitivity equivalent to that of an expert, with a better balance between sensitivity and specificity, which implies that AI can play an important role in primary diagnosis of TMJOA from OPGs in most general practice clinics where OMFR experts or CT are not available.

Advancements in Dentistry with Artificial Intelligence: Current Clinical Applications and Future Perspectives

“…Moreover, a deep CNN (ResNet) was presented by Choi et al to diagnose temporomandibular joint disorders (TMJ) osteoarthritis. The model yielded an accuracy of 78% similar to that of Oral and maxillofacial radiologists (OMFR) [ 58 ].…”

“…The limitations of each relevant study covered in this review are given in Table 8 and Table 9 . One of the most prevailing limitation in majority of the studies involving application of AI in subfields of dentistry is dataset limitation in terms of size for disease diagnosis [ 5 , 19 , 31 , 42 , 44 , 45 , 51 , 56 , 57 , 58 , 59 , 71 ], for treatment planning and prognosis [ 29 , 37 , 62 ], and for landmark detection [ 64 ]. Other limitations include relying on the reliability of examiners for accurate diagnosis [ 41 ], less flexibility to capture nonlinearities in data [ 23 ], limited practical use [ 54 ], poor performance on images that are not evenly illuminated [ 6 ], limited class discrimination [ 6 ], lack in terms of generalizability [ 29 ], lack in terms of performance [ 9 , 28 , 60 , 61 , 62 , 64 , 65 , 66 , 68 , 69 ], and computational capabilities [ 67 ].…”

Artificial intelligence in orthodontics: A way towards modernization

“…To eliminate this problem an AI algorithm was developed and trained to read TMJ osteoarthritis on OPGs. 20…”

Artificial intelligence to support early diagnosis of temporomandibular disorders: A preliminary case study

“…9 In fact, in the last 2 years, ML-AI techniques, which are extensively used in healthcare, faced an increasing adoption also in dentistry, 9 and more specifically in TMD diagnosis. [10][11][12][13][14] As ML requires large datasets for appropriate training, the literature mostly reports the use of automated techniques for the interpretation of bioimages, mostly Computerised Tomography (CT) or Magnetic Resonance Imaging scans. [10][11][12][13][14][15] These are used in TMD diagnosis as quantitative tools to appropriately classify the type of disk displacement, with high accuracy.…”

“…As ML requires large datasets for appropriate training, the literature mostly reports the use of automated techniques for the interpretation of bioimages, mostly Computerised Tomography (CT) or Magnetic Resonance Imaging scans 10–15 . These are used in TMD diagnosis as quantitative tools to appropriately classify the type of disk displacement, with high accuracy.…”