Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images
Background Previous automated models inaccurately scored non-standardized plaque images. The objectives were to develop and test automated image selection and intraoral plaque-scoring (primary outcome measure in a prevention trial for preschoolers).Methods Evaluating 1650 plaque-disclosed primary te...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | Journal of the California Dental Association |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/19424396.2024.2422146 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850131464683782144 |
|---|---|
| author | Eric Coy William Santo Bonnie Jue Helen Betts Francisco Ramos-Gomez Stuart A. Gansky |
| author_facet | Eric Coy William Santo Bonnie Jue Helen Betts Francisco Ramos-Gomez Stuart A. Gansky |
| author_sort | Eric Coy |
| collection | DOAJ |
| description | Background Previous automated models inaccurately scored non-standardized plaque images. The objectives were to develop and test automated image selection and intraoral plaque-scoring (primary outcome measure in a prevention trial for preschoolers).Methods Evaluating 1650 plaque-disclosed primary teeth (teeth D, E, F, G) from 435 photographs from UCSF/UCLA clinical trials, data were cleaned, transformed, and modeled with statistical and machine learning (ML) algorithms; data visualizations utilized Jupyter Notebooks, Python, OpenCV, and Sci-kit Learn libraries, with Laplacian filter preprocessing. Image selection and plaque-scoring used 8 ML classification models. Mean plaque-scoring used 8 ML regression models. Models were tuned with 80:20 train:test split, stratified 5-fold cross-validation (5-CV) (unstratified in regression models), and hyperparameter optimization. Area-under-the-curve receiver operating characteristic (AUC-ROC) curve and R2 determined the best classification and regression models, respectively, compared to calibrated dentist researcher ratings. Training time was a secondary metric. Manual segmentation used Photoshop’s lasso tool. Average and dominant hue, saturation, and brightness values were features for training plaque-scoring algorithms.Results Best performing models were: Support Vector Machine-Gaussian for image selection, 5-CV AUC-ROC of 0.99 and 0.76s of training time; Gradient-Boosting classification and regression models for individual teeth (5-CV AUC-ROC of 0.99 with 105s training); and mean plaque-scoring algorithms (5-CV R2 of 0.72 with 1415s training).Conclusions Accurate automated plaque-scoring is attainable without the high computational and financial costs of deep learning (DL) models. Automated plaque-scoring is attainable with little user-manipulation.Practical Implications Implementing automated tooth segmentation and synthetic sample generation with DL training may strengthen reliability, validity, and efficiency for clinical, research, and teledentistry applications by eliminating manual image preprocessing. |
| format | Article |
| id | doaj-art-1a16447e158742e0b21f03ec80f0a142 |
| institution | OA Journals |
| issn | 1942-4396 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Journal of the California Dental Association |
| spelling | doaj-art-1a16447e158742e0b21f03ec80f0a1422025-08-20T02:32:25ZengTaylor & Francis GroupJournal of the California Dental Association1942-43962024-12-0152110.1080/19424396.2024.2422146Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized ImagesEric Coy0William Santo1Bonnie Jue2Helen Betts3Francisco Ramos-Gomez4Stuart A. Gansky5University of California San Francisco School of Dentistry, San Francisco, Califonia, USAUCSF Bakar Computational Health Sciences Institute, UCSF School of Dentistry, San Francisco, Califonia, USAUniversity of California San Francisco School of Dentistry, San Francisco, Califonia, USAUniversity of California Los Angeles School of Dentistry, Los Angeles, Califonia, USAUniversity of California Los Angeles School of Dentistry, Los Angeles, Califonia, USAUniversity of California San Francisco School of Dentistry, San Francisco, Califonia, USABackground Previous automated models inaccurately scored non-standardized plaque images. The objectives were to develop and test automated image selection and intraoral plaque-scoring (primary outcome measure in a prevention trial for preschoolers).Methods Evaluating 1650 plaque-disclosed primary teeth (teeth D, E, F, G) from 435 photographs from UCSF/UCLA clinical trials, data were cleaned, transformed, and modeled with statistical and machine learning (ML) algorithms; data visualizations utilized Jupyter Notebooks, Python, OpenCV, and Sci-kit Learn libraries, with Laplacian filter preprocessing. Image selection and plaque-scoring used 8 ML classification models. Mean plaque-scoring used 8 ML regression models. Models were tuned with 80:20 train:test split, stratified 5-fold cross-validation (5-CV) (unstratified in regression models), and hyperparameter optimization. Area-under-the-curve receiver operating characteristic (AUC-ROC) curve and R2 determined the best classification and regression models, respectively, compared to calibrated dentist researcher ratings. Training time was a secondary metric. Manual segmentation used Photoshop’s lasso tool. Average and dominant hue, saturation, and brightness values were features for training plaque-scoring algorithms.Results Best performing models were: Support Vector Machine-Gaussian for image selection, 5-CV AUC-ROC of 0.99 and 0.76s of training time; Gradient-Boosting classification and regression models for individual teeth (5-CV AUC-ROC of 0.99 with 105s training); and mean plaque-scoring algorithms (5-CV R2 of 0.72 with 1415s training).Conclusions Accurate automated plaque-scoring is attainable without the high computational and financial costs of deep learning (DL) models. Automated plaque-scoring is attainable with little user-manipulation.Practical Implications Implementing automated tooth segmentation and synthetic sample generation with DL training may strengthen reliability, validity, and efficiency for clinical, research, and teledentistry applications by eliminating manual image preprocessing.https://www.tandfonline.com/doi/10.1080/19424396.2024.2422146Artificial intelligence/classificationautomated pattern recognition/classificationdental photography/classificationdental plaque index/diagnostic imagingmachine learning/classificationobserver variation |
| spellingShingle | Eric Coy William Santo Bonnie Jue Helen Betts Francisco Ramos-Gomez Stuart A. Gansky Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images Journal of the California Dental Association Artificial intelligence/classification automated pattern recognition/classification dental photography/classification dental plaque index/diagnostic imaging machine learning/classification observer variation |
| title | Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images |
| title_full | Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images |
| title_fullStr | Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images |
| title_full_unstemmed | Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images |
| title_short | Among Artificial Intelligence/Machine Learning Methods, Automated Gradient-Boosting Models Accurately Score Intraoral Plaque in Non-Standardized Images |
| title_sort | among artificial intelligence machine learning methods automated gradient boosting models accurately score intraoral plaque in non standardized images |
| topic | Artificial intelligence/classification automated pattern recognition/classification dental photography/classification dental plaque index/diagnostic imaging machine learning/classification observer variation |
| url | https://www.tandfonline.com/doi/10.1080/19424396.2024.2422146 |
| work_keys_str_mv | AT ericcoy amongartificialintelligencemachinelearningmethodsautomatedgradientboostingmodelsaccuratelyscoreintraoralplaqueinnonstandardizedimages AT williamsanto amongartificialintelligencemachinelearningmethodsautomatedgradientboostingmodelsaccuratelyscoreintraoralplaqueinnonstandardizedimages AT bonniejue amongartificialintelligencemachinelearningmethodsautomatedgradientboostingmodelsaccuratelyscoreintraoralplaqueinnonstandardizedimages AT helenbetts amongartificialintelligencemachinelearningmethodsautomatedgradientboostingmodelsaccuratelyscoreintraoralplaqueinnonstandardizedimages AT franciscoramosgomez amongartificialintelligencemachinelearningmethodsautomatedgradientboostingmodelsaccuratelyscoreintraoralplaqueinnonstandardizedimages AT stuartagansky amongartificialintelligencemachinelearningmethodsautomatedgradientboostingmodelsaccuratelyscoreintraoralplaqueinnonstandardizedimages |