Publications
2023
F, Miranda; V, Choudhari; S, Barone; L, Anchling; N, Hutin; M, Gurgel; et al,
Interpretable artificial intelligence for classification of alveolar bone defect in patients with cleft lip and palate. Journal Article
In: Scientific Reports, vol. 15861, 2023.
Abstract | Links | BibTeX | Tags: 3D landmark identification, alveolar bone defect, artificial intelligence, cleft lip, cleft lip and palate
@article{Bianchi2023j,
title = {Interpretable artificial intelligence for classification of alveolar bone defect in patients with cleft lip and palate. },
author = {Miranda F and Choudhari V and Barone S and Anchling L and Hutin N and Gurgel M and et al},
url = {https://doi.org/10.1038/s41598-023-43125-7},
doi = {10.1038/s41598-023-43125-7},
year = {2023},
date = {2023-09-22},
journal = {Scientific Reports},
volume = {15861},
abstract = {Cleft lip and/or palate (CLP) is the most common congenital craniofacial anomaly and requires bone grafting of the alveolar cleft. This study aimed to develop a novel classification algorithm to assess the severity of alveolar bone defects in patients with CLP using three-dimensional (3D) surface models and to demonstrate through an interpretable artificial intelligence (AI)-based algorithm the decisions provided by the classifier. Cone-beam computed tomography scans of 194 patients with CLP were used to train and test the performance of an automatic classification of the severity of alveolar bone defect. The shape, height, and width of the alveolar bone defect were assessed in automatically segmented maxillary 3D surface models to determine the ground truth classification index of its severity. The novel classifier algorithm renders the 3D surface models from different viewpoints and captures 2D image snapshots fed into a 2D Convolutional Neural Network. An interpretable AI algorithm was developed that uses features from each view and aggregated via Attention Layers to explain the classification. The precision, recall and F-1 score were 0.823, 0.816, and 0.817, respectively, with agreement ranging from 97.4 to 100% on the severity index within 1 group difference. The new classifier and interpretable AI algorithm presented satisfactory accuracy to classify the severity of alveolar bone defect morphology using 3D surface models of patients with CLP and graphically displaying the features that were considered during the deep learning model's classification decision.},
keywords = {3D landmark identification, alveolar bone defect, artificial intelligence, cleft lip, cleft lip and palate},
pubstate = {published},
tppubtype = {article}
}
Cleft lip and/or palate (CLP) is the most common congenital craniofacial anomaly and requires bone grafting of the alveolar cleft. This study aimed to develop a novel classification algorithm to assess the severity of alveolar bone defects in patients with CLP using three-dimensional (3D) surface models and to demonstrate through an interpretable artificial intelligence (AI)-based algorithm the decisions provided by the classifier. Cone-beam computed tomography scans of 194 patients with CLP were used to train and test the performance of an automatic classification of the severity of alveolar bone defect. The shape, height, and width of the alveolar bone defect were assessed in automatically segmented maxillary 3D surface models to determine the ground truth classification index of its severity. The novel classifier algorithm renders the 3D surface models from different viewpoints and captures 2D image snapshots fed into a 2D Convolutional Neural Network. An interpretable AI algorithm was developed that uses features from each view and aggregated via Attention Layers to explain the classification. The precision, recall and F-1 score were 0.823, 0.816, and 0.817, respectively, with agreement ranging from 97.4 to 100% on the severity index within 1 group difference. The new classifier and interpretable AI algorithm presented satisfactory accuracy to classify the severity of alveolar bone defect morphology using 3D surface models of patients with CLP and graphically displaying the features that were considered during the deep learning model's classification decision.
2022
Ghowsi, A; Hatcher, D; Suh, H; Wiled, D; Castro, W; Krueger, J; Park, J; Oh, H
Automated landmark identification on one cone beam computed tomography: Accuracy and reliability Journal Article
In: Angle Orthodontist, vol. 92, pp. 642-654, 2022.
Abstract | Links | BibTeX | Tags: 3D landmark identification, AAOF, accuracy, Automated, CBCT, Landmark error, Reliability
@article{Oh2022b,
title = {Automated landmark identification on one cone beam computed tomography: Accuracy and reliability},
author = {A Ghowsi and D Hatcher and H Suh and D Wiled and W Castro and J Krueger and J Park and H Oh},
url = {https://pubmed.ncbi.nlm.nih.gov/35653226/},
doi = {10.2319/122121-928.1},
year = {2022},
date = {2022-06-02},
urldate = {2022-06-02},
journal = {Angle Orthodontist},
volume = {92},
pages = {642-654},
abstract = {Objectives: To evaluate the accuracy and reliability of a fully automated landmark identification (ALI) system as a tool for automatic landmark location compared with human judges.
Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated.
Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range.
Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs.},
keywords = {3D landmark identification, AAOF, accuracy, Automated, CBCT, Landmark error, Reliability},
pubstate = {published},
tppubtype = {article}
}
Objectives: To evaluate the accuracy and reliability of a fully automated landmark identification (ALI) system as a tool for automatic landmark location compared with human judges.
Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated.
Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range.
Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs.
Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated.
Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range.
Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs.
F, Miranda; V, Choudhari; S, Barone; L, Anchling; N, Hutin; M, Gurgel; et al,
Interpretable artificial intelligence for classification of alveolar bone defect in patients with cleft lip and palate. Journal Article
In: Scientific Reports, vol. 15861, 2023.
@article{Bianchi2023j,
title = {Interpretable artificial intelligence for classification of alveolar bone defect in patients with cleft lip and palate. },
author = {Miranda F and Choudhari V and Barone S and Anchling L and Hutin N and Gurgel M and et al},
url = {https://doi.org/10.1038/s41598-023-43125-7},
doi = {10.1038/s41598-023-43125-7},
year = {2023},
date = {2023-09-22},
journal = {Scientific Reports},
volume = {15861},
abstract = {Cleft lip and/or palate (CLP) is the most common congenital craniofacial anomaly and requires bone grafting of the alveolar cleft. This study aimed to develop a novel classification algorithm to assess the severity of alveolar bone defects in patients with CLP using three-dimensional (3D) surface models and to demonstrate through an interpretable artificial intelligence (AI)-based algorithm the decisions provided by the classifier. Cone-beam computed tomography scans of 194 patients with CLP were used to train and test the performance of an automatic classification of the severity of alveolar bone defect. The shape, height, and width of the alveolar bone defect were assessed in automatically segmented maxillary 3D surface models to determine the ground truth classification index of its severity. The novel classifier algorithm renders the 3D surface models from different viewpoints and captures 2D image snapshots fed into a 2D Convolutional Neural Network. An interpretable AI algorithm was developed that uses features from each view and aggregated via Attention Layers to explain the classification. The precision, recall and F-1 score were 0.823, 0.816, and 0.817, respectively, with agreement ranging from 97.4 to 100% on the severity index within 1 group difference. The new classifier and interpretable AI algorithm presented satisfactory accuracy to classify the severity of alveolar bone defect morphology using 3D surface models of patients with CLP and graphically displaying the features that were considered during the deep learning model's classification decision.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Cleft lip and/or palate (CLP) is the most common congenital craniofacial anomaly and requires bone grafting of the alveolar cleft. This study aimed to develop a novel classification algorithm to assess the severity of alveolar bone defects in patients with CLP using three-dimensional (3D) surface models and to demonstrate through an interpretable artificial intelligence (AI)-based algorithm the decisions provided by the classifier. Cone-beam computed tomography scans of 194 patients with CLP were used to train and test the performance of an automatic classification of the severity of alveolar bone defect. The shape, height, and width of the alveolar bone defect were assessed in automatically segmented maxillary 3D surface models to determine the ground truth classification index of its severity. The novel classifier algorithm renders the 3D surface models from different viewpoints and captures 2D image snapshots fed into a 2D Convolutional Neural Network. An interpretable AI algorithm was developed that uses features from each view and aggregated via Attention Layers to explain the classification. The precision, recall and F-1 score were 0.823, 0.816, and 0.817, respectively, with agreement ranging from 97.4 to 100% on the severity index within 1 group difference. The new classifier and interpretable AI algorithm presented satisfactory accuracy to classify the severity of alveolar bone defect morphology using 3D surface models of patients with CLP and graphically displaying the features that were considered during the deep learning model's classification decision.
Ghowsi, A; Hatcher, D; Suh, H; Wiled, D; Castro, W; Krueger, J; Park, J; Oh, H
Automated landmark identification on one cone beam computed tomography: Accuracy and reliability Journal Article
In: Angle Orthodontist, vol. 92, pp. 642-654, 2022.
@article{Oh2022b,
title = {Automated landmark identification on one cone beam computed tomography: Accuracy and reliability},
author = {A Ghowsi and D Hatcher and H Suh and D Wiled and W Castro and J Krueger and J Park and H Oh},
url = {https://pubmed.ncbi.nlm.nih.gov/35653226/},
doi = {10.2319/122121-928.1},
year = {2022},
date = {2022-06-02},
urldate = {2022-06-02},
journal = {Angle Orthodontist},
volume = {92},
pages = {642-654},
abstract = {Objectives: To evaluate the accuracy and reliability of a fully automated landmark identification (ALI) system as a tool for automatic landmark location compared with human judges.
Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated.
Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range.
Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Objectives: To evaluate the accuracy and reliability of a fully automated landmark identification (ALI) system as a tool for automatic landmark location compared with human judges.
Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated.
Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range.
Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs.
Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated.
Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range.
Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs.
2023 |
F, Miranda; V, Choudhari; S, Barone; L, Anchling; N, Hutin; M, Gurgel; et al,: Interpretable artificial intelligence for classification of alveolar bone defect in patients with cleft lip and palate. . In: Scientific Reports, vol. 15861, 2023. (Type: Journal Article | Abstract | Links | BibTeX | Tags: 3D landmark identification, alveolar bone defect, artificial intelligence, cleft lip, cleft lip and palate)@article{Bianchi2023j,Cleft lip and/or palate (CLP) is the most common congenital craniofacial anomaly and requires bone grafting of the alveolar cleft. This study aimed to develop a novel classification algorithm to assess the severity of alveolar bone defects in patients with CLP using three-dimensional (3D) surface models and to demonstrate through an interpretable artificial intelligence (AI)-based algorithm the decisions provided by the classifier. Cone-beam computed tomography scans of 194 patients with CLP were used to train and test the performance of an automatic classification of the severity of alveolar bone defect. The shape, height, and width of the alveolar bone defect were assessed in automatically segmented maxillary 3D surface models to determine the ground truth classification index of its severity. The novel classifier algorithm renders the 3D surface models from different viewpoints and captures 2D image snapshots fed into a 2D Convolutional Neural Network. An interpretable AI algorithm was developed that uses features from each view and aggregated via Attention Layers to explain the classification. The precision, recall and F-1 score were 0.823, 0.816, and 0.817, respectively, with agreement ranging from 97.4 to 100% on the severity index within 1 group difference. The new classifier and interpretable AI algorithm presented satisfactory accuracy to classify the severity of alveolar bone defect morphology using 3D surface models of patients with CLP and graphically displaying the features that were considered during the deep learning model's classification decision. |
2022 |
Ghowsi, A; Hatcher, D; Suh, H; Wiled, D; Castro, W; Krueger, J; Park, J; Oh, H: Automated landmark identification on one cone beam computed tomography: Accuracy and reliability. In: Angle Orthodontist, vol. 92, pp. 642-654, 2022. (Type: Journal Article | Abstract | Links | BibTeX | Tags: 3D landmark identification, AAOF, accuracy, Automated, CBCT, Landmark error, Reliability)@article{Oh2022b,Objectives: To evaluate the accuracy and reliability of a fully automated landmark identification (ALI) system as a tool for automatic landmark location compared with human judges. Materials and methods: A total of 100 cone-beam computed tomography (CBCT) images were collected. After the calibration procedure, two human judges identified 53 landmarks in the x, y, and z coordinate planes on CBCTs using Checkpoint Software (Stratovan Corporation, Davis, Calif). The ground truth was created by averaging landmark coordinates identified by two human judges for each landmark. To evaluate the accuracy of ALI, the mean absolute error (mm) at the x, y, and z coordinates and mean error distance (mm) between the human landmark identification and the ALI were determined, and a successful detection rate was calculated. Results: Overall, the ALI system was as successful at landmarking as the human judges. The ALI's mean absolute error for all coordinates was 1.57 mm on average. Across all three coordinate planes, 94% of the landmarks had a mean absolute error of less than 3 mm. The mean error distance for all 53 landmarks was 3.19 ± 2.6 mm. When applied to 53 landmarks on 100 CBCTs, the ALI system showed a 75% success rate in detecting landmarks within a 4-mm error distance range. Conclusions: Overall, ALI showed clinically acceptable mean error distances except for a few landmarks. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identifications on CBCTs. |