Primary study · included by 2 reviews

Can temporomandibular joint osteoarthritis be diagnosed on MRI proton density-weighted images with diagnostic support from the latest deep learning classification models?

Nozawa 2025 · Nozawa Michihito, Fukuda Motoki, Kotaki Shinya, et al. · Dento maxillo facial radiology · 2025

Included by

Non-Cochrane · DTA · 2026 · cited as Nozawa 2024 · Reference list [23]; Table 1; Figs. 3, 5, 6

Outcome this study informs	Pooled effect	Evidence	Certainty
Specificity AI (machine/deep learning) models, all architectures pooled vs Reference standard diagnosis of TMJ anomaly	Specificity (pooled proportion) 0.399 [0.343, 0.456]	5 studies	GRADE not reported
Accuracy AI (machine/deep learning) models, all architectures pooled vs Reference standard diagnosis of TMJ anomaly	Accuracy (pooled proportion) 0.487 [0.403, 0.571]	3 studies	GRADE not reported
Sensitivity AI (machine/deep learning) models, all architectures pooled vs Reference standard diagnosis of TMJ anomaly	Sensitivity (pooled proportion) 0.399 [0.348, 0.450]	6 studies	GRADE not reported

Non-Cochrane · DTA · 2026 · cited as Nozawa 2025 · Table 3; reference [17]

Outcome this study informs	Pooled effect	Evidence	Certainty
Diagnostic accuracy: sensitivity AI-based image classification (deep learning) for TMJ osteoarthritis vs Reference standard (CBCT or CT-supported expert radiological diagnosis)	Sensitivity 0.900	7 studies	GRADE not reported
Diagnostic accuracy: sensitivity AI-based image classification (deep learning) for TMJ osteoarthritis vs Reference standard (CBCT or CT-supported expert radiological diagnosis)	Sensitivity 0.791 [0.700, 0.861]	3 studies	GRADE not reported
Diagnostic accuracy: sensitivity AI-based image classification (deep learning) for TMJ osteoarthritis vs Reference standard (CBCT or CT-supported expert radiological diagnosis)	Sensitivity 0.748 [0.688, 0.801]	2 studies	GRADE not reported

2 including reviews in total.