Kim-Cuong T Nguyen1, Binh M Le2, Mengxun Li3, Fabiana T Almeida4, Paul W Major4, Neelambar R Kaipatur4, Edmond H M Lou5, Kumaradevan Punithakumar6, Lawrence H Le7. 1. Department of Radiology & Diagnostic Imaging, University of Alberta, Edmonton, Canada; Department of Biomedical Engineering, University of Alberta, Edmonton, Canada. 2. Department of Radiology & Diagnostic Imaging, University of Alberta, Edmonton, Canada; Department of Computer Sciences, University of Science, Ho Chi Minh City, Vietnam. 3. Department of Implantology, School and Hospital of Stomatology, Wuhan University, Wuhan, China. 4. School of Dentistry, University of Alberta, Edmonton, Canada. 5. Department of Biomedical Engineering, University of Alberta, Edmonton, Canada; Department of Electrical & Computer Engineering, University of Alberta, Edmonton, Canada. 6. Department of Radiology & Diagnostic Imaging, University of Alberta, Edmonton, Canada. Electronic address: punithak@ualberta.ca. 7. Department of Radiology & Diagnostic Imaging, University of Alberta, Edmonton, Canada; Department of Biomedical Engineering, University of Alberta, Edmonton, Canada; School of Dentistry, University of Alberta, Edmonton, Canada. Electronic address: awrence.le@ualberta.ca.
Abstract
OBJECTIVE: Our goal was to automatically identify the cementoenamel junction (CEJ) location in ultrasound images using deep convolution neural networks (CNNs). METHODS: Three CNNs were evaluated using 1400 images and data augmentation. The training and validation were performed by an experienced nonclinical rater with 1000 and 200 images, respectively. Four clinical raters with different levels of experience with ultrasound tested the networks using the other 200 images. In addition to the comparison of the best approach with each rater, we also employed the simultaneous truth and performance level estimation (STAPLE) algorithm to estimate a ground truth based on all labelings by four clinical raters. The final CEJ location estimate was obtained by taking the first moment of the posterior probability computed using the STAPLE algorithm. The study also computes the machine learning-measured CEJ-alveolar bone crest distance. RESULTS: Quantitative evaluations of the 200 images showed that the comparison of the best approach with the STAPLE-estimate yielded a mean difference (MD) of 0.26 mm, which is close to the comparison with the most experienced nonclinical rater (MD=0.25 mm) but far better than the comparison with clinical raters (MD=0.27-0.33 mm). The machine learning-measured CEJ-alveolar bone crest distances correlated strongly (R = 0.933, p < 0.001) with the manual clinical labeling and the measurements were in good agreement with the 95% Bland-Altman's lines of agreement between -0.68 and 0.57 mm. CONCLUSIONS: The study demonstrated the feasible use of machine learning methodology to localize CEJ in ultrasound images with clinically acceptable accuracy and reliability. Likelihood-weighted ground truth by combining multiple labels by the clinical experts compared favorably with the predictions by the best deep CNN approach. CLINICAL SIGNIFICANCE: Identification of CEJ and its distance from the alveolar bone crest play an important role in the evaluation of periodontal status. Machine learning algorithms can learn from complex features in ultrasound images and have potential to provide a reliable and accurate identification in subsecond. This will greatly assist dental practitioners to provide better point-of-care to patients and enhance the throughput of dental care.
OBJECTIVE: Our goal was to automatically identify the cementoenamel junction (CEJ) location in ultrasound images using deep convolution neural networks (CNNs). METHODS: Three CNNs were evaluated using 1400 images and data augmentation. The training and validation were performed by an experienced nonclinical rater with 1000 and 200 images, respectively. Four clinical raters with different levels of experience with ultrasound tested the networks using the other 200 images. In addition to the comparison of the best approach with each rater, we also employed the simultaneous truth and performance level estimation (STAPLE) algorithm to estimate a ground truth based on all labelings by four clinical raters. The final CEJ location estimate was obtained by taking the first moment of the posterior probability computed using the STAPLE algorithm. The study also computes the machine learning-measured CEJ-alveolar bone crest distance. RESULTS: Quantitative evaluations of the 200 images showed that the comparison of the best approach with the STAPLE-estimate yielded a mean difference (MD) of 0.26 mm, which is close to the comparison with the most experienced nonclinical rater (MD=0.25 mm) but far better than the comparison with clinical raters (MD=0.27-0.33 mm). The machine learning-measured CEJ-alveolar bone crest distances correlated strongly (R = 0.933, p < 0.001) with the manual clinical labeling and the measurements were in good agreement with the 95% Bland-Altman's lines of agreement between -0.68 and 0.57 mm. CONCLUSIONS: The study demonstrated the feasible use of machine learning methodology to localize CEJ in ultrasound images with clinically acceptable accuracy and reliability. Likelihood-weighted ground truth by combining multiple labels by the clinical experts compared favorably with the predictions by the best deep CNN approach. CLINICAL SIGNIFICANCE: Identification of CEJ and its distance from the alveolar bone crest play an important role in the evaluation of periodontal status. Machine learning algorithms can learn from complex features in ultrasound images and have potential to provide a reliable and accurate identification in subsecond. This will greatly assist dental practitioners to provide better point-of-care to patients and enhance the throughput of dental care.
Authors: Radu Chifor; Mircea Hotoleanu; Tiberiu Marita; Tudor Arsenescu; Mihai Adrian Socaciu; Iulia Clara Badea; Ioana Chifor Journal: Sensors (Basel) Date: 2022-09-20 Impact factor: 3.847