Primary study · included by 1 review

Automatic three-dimensional nasal and pharyngeal airway subregions identification via Vision Transformer.

Jin 2023 · Jin Suhan, Han Haojie, Huang Zhiqun, et al. · Journal of dentistry · 2023

Included by

Outcome this study informsPooled effectEvidenceCertainty
Recall (sensitivity) of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.92 [0.87, 0.97]3 studiesGRADE not reported
Dice similarity score
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.94 [0.91, 0.97]3 studiesGRADE not reported
Precision of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.95 [0.92, 0.97]3 studiesGRADE not reported
Precision of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.95 [0.93, 0.97]4 studiesGRADE not reported
Dice similarity score
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.95 [0.92, 0.98]3 studiesGRADE not reported
Precision of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.96 [0.93, 0.99]3 studiesGRADE not reported
Intersection over union (IoU)
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.93 [0.92, 0.94]1 studyGRADE not reported
Recall (sensitivity)
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.95 [0.92, 0.99]3 studiesGRADE not reported
Recall (sensitivity)
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.93 [0.87, 0.99]3 studiesGRADE not reported
Recall (sensitivity) of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.94 [0.89, 0.98]4 studiesGRADE not reported
Recall (sensitivity) of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.94 [0.88, 1.0]3 studiesGRADE not reported
Precision
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.96 [0.94, 0.98]3 studiesGRADE not reported
Dice similarity score
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.95 [0.93, 0.98]3 studiesGRADE not reported
Dice similarity score
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.94 [0.92, 0.97]4 studiesGRADE not reported
Dice similarity score
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.94 [0.91, 0.96]3 studiesGRADE not reported
Intersection over union (IoU)
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.93 [0.92, 0.94]1 studyGRADE not reported
Intersection over union (IoU) of AI segmentation
AI-based (deep learning or machine learning) 3D segmentation vs Manual segmentation by human experts
Weighted mean 0.93 [0.92, 0.94]2 studiesGRADE not reported
Precision
Automatic 3D segmentation by artificial intelligence (deep learning or machine learning) vs Manual segmentation by human experts
Weighted mean 0.95 [0.92, 0.97]3 studiesGRADE not reported

1 including review in total.