Kodak Dental Imaging Software 6 7 11

Download ✒ ✒ ✒ DOWNLOAD


Kodak Dental Imaging Software 6 7 11

in this study, the sensitivity of the system was 92% for the detection of periodontal disease, which was slightly better than the sensitivity of human examiners that ranged from 89% to 92% for the same condition. the system also detected the presence or absence of signs of dental caries with a sensitivity of 97% and specificity of 96%. the diagnocat system could correctly detect dental endodontic treatment in 100% of the cases, with a sensitivity of 100% and a specificity of 92%. however, the system could not detect the nature of the periodontal disease, which was unclear from the images in the dataset. the system also could not identify the presence or absence of a root canal. this was presumably due to the fact that root canal treatment information is not incorporated into the training data and it is not directly detected by the system.

the ai system was not able to detect the presence or absence of dental caries in the tooth structure. this was probably due to the absence of caries information in the training data. however, the system correctly identified periodontal disease with a sensitivity of 97% and a specificity of 95%. the system also identified the presence or absence of a tooth with a sensitivity of 91% and a specificity of 99%.

as mentioned, this paper is the first to evaluate the diagnostic performance of the diagnocat software. there were no differences in the performance of the human examiners and the ai system in terms of both sensitivity and specificity. however, the precision and accuracy of the system were higher for most conditions, and, importantly, when more than one condition was present. this is in accordance with previous studies, which have shown that ai is capable of providing more accurate results in multiple conditions. 20, 8, 10, 21 thus, the ai system could have helped in the diagnosis and treatment planning of cases that were difficult to diagnose by the dental radiology examiners. in addition, the ai system could have helped in the diagnosis of a number of endodontic conditions such as post treatment areas, missed root canals, and apical periodontitis. we believe that the accuracy of the diagnosis provided by the ai system may improve further with additional post processing and training. for example, the background provided by the diagnocat software could be used as a feature for more advanced machine learning algorithms to learn and categorize endodontic disease in a way that is difficult for human observers. moreover, this ai system could be used to train the human operators of the dental radiology examiners in recognizing these challenging situations and, thus, increasing their sensitivity and specificity for the diagnosis of endodontic conditions in cbct images.