Christopher G. Anton
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Christopher G. Anton has a total of 58 co-authors. Here are the top co-authors by number of shared publications:
- Andrew Timothy Trout
Department of Radiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, USA
5 shared publications - Alexander J. Towbin
Department of Radiology Cincinnati Children's Hospital Medical Center Cincinnati Ohio USA
3 shared publications - Jonathan R. Dillman
Radiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, USA
3 shared publications - Susan E. Sharp
Cincinnati Children's Hospital Medical Center Department of Radiology, United States (M.G.H.K., C.E.M., S.E.S., A.T.T.); University of Cincinnati College of Medicine Department of Radiology, United States (C.E.M., S.E.S., A.T.T.).
3 shared publications - Andi Li
Department of Biomedical Engineering, University of Cincinnati, Cincinnati, OH, USA
2 shared publications - Chenyang Han
Department of Biomedical Engineering, University of Cincinnati, 2901 Woodside Drive, Cincinnati, OH, 45219, USA
2 shared publications
Latest Publications
- Open Access
- 3 November 2026
- Published by Elsevier in NeuroImage
NA-DIP: A Noise-Aware Deep Image Prior for Generating Dedicated Brain PET–Quality Images from Whole-Body PET Acquisitions
- 3 November 2026
- Published by Elsevier in NeuroImage
Open AccessNA-DIP: A Noise-Aware Deep Image Prior for Generating Dedicated Brain PET–Quality Images from Whole-Body PET Acquisitions
- Open Access
- 21 May 2026
- Published by Springer Nature in EJNMMI Physics
- Vol. 13 (1)
Acquisition time/dose reduction in pediatric PET imaging using patch-based deep learning
Deep learning (DL)-based denoising methods have shown promise for reducing radiation dose and/or acquisition time in pediatric PET imaging. However, conventional DL approaches typically require large and diverse training datasets to achieve generalizability. We propose a patch-based DL (PDL) approach that learns local structural representations from a single high-quality examination, with the goal of improving image quality and preserving quantitative accuracy of reduced-count pediatric whole-body PET while minimizing training burden. This retrospective study included PET/CT datasets acquired on a GE Discovery MI Gen 2 scanner. A single pediatric examination with high structural conspicuity and low noise was used for training. Testing was performed using a NEMA phantom and 88 clinical pediatric examinations (age < 12 years) acquired with routine pediatric FDG dosing and a 90‑s‑per‑bed protocol. Clinical list‑mode data were truncated to 20 s per bed to simulate reduced‑count acquisitions; training and phantom datasets were downsampled to match this noise level. The PDL model was trained using paired full‑count and reduced‑count image patches and applied to denoise reduced‑count phantom and patient images. Phantom evaluations assessed spatial resolution, contrast recovery, and noise. Clinical evaluation included a blinded observer study and quantitative SUV analysis, followed by assessment of the noise-bias tradeoff. PDL preserved spatial resolution and contrast while achieving substantial noise reduction in phantom experiments. In the observer study, PDL‑enhanced 20‑s‑per‑bed images were rated as equivalent or superior in overall image quality and noise relative to standard 90‑s‑per‑bed images in more than 85% of cases. Small negative relative bias in SUVmax and SUVmean was observed in the liver, mediastinal blood pool, and lesions, with no statistically significant differences compared with standard‑count images. Compared with highly regularized reconstruction, PDL achieved greater noise reduction with reduced bias. PDL enhancement for pediatric whole-body PET imaging may enable a reduction in acquisition time to 20 s per bed or a reduction in administered activity to approximately 22% of current clinical practice, while maintaining spatial resolution, contrast, quantitative accuracy, and diagnostic image quality.
- 21 May 2026
- Published by Springer Nature in EJNMMI Physics
- Vol. 13 (1)
Open AccessAcquisition time/dose reduction in pediatric PET imaging using patch-based deep learning
Deep learning (DL)-based denoising methods have shown promise for reducing radiation dose and/or acquisition time in pediatric PET imaging. However, conventional DL approaches typically require large and diverse training datasets to achieve generalizability. We propose a patch-based DL (PDL) approach that learns local structural representations from a single high-quality examination, with the goal of improving image quality and preserving quantitative accuracy of reduced-count pediatric whole-body PET while minimizing training burden. This retrospective study included PET/CT datasets acquired on a GE Discovery MI Gen 2 scanner. A single pediatric examination with high structural conspicuity and low noise was used for training. Testing was performed using a NEMA phantom and 88 clinical pediatric examinations (age < 12 years) acquired with routine pediatric FDG dosing and a 90‑s‑per‑bed protocol. Clinical list‑mode data were truncated to 20 s per bed to simulate reduced‑count acquisitions; training and phantom datasets were downsampled to match this noise level. The PDL model was trained using paired full‑count and reduced‑count image patches and applied to denoise reduced‑count phantom and patient images. Phantom evaluations assessed spatial resolution, contrast recovery, and noise. Clinical evaluation included a blinded observer study and quantitative SUV analysis, followed by assessment of the noise-bias tradeoff. PDL preserved spatial resolution and contrast while achieving substantial noise reduction in phantom experiments. In the observer study, PDL‑enhanced 20‑s‑per‑bed images were rated as equivalent or superior in overall image quality and noise relative to standard 90‑s‑per‑bed images in more than 85% of cases. Small negative relative bias in SUVmax and SUVmean was observed in the liver, mediastinal blood pool, and lesions, with no statistically significant differences compared with standard‑count images. Compared with highly regularized reconstruction, PDL achieved greater noise reduction with reduced bias. PDL enhancement for pediatric whole-body PET imaging may enable a reduction in acquisition time to 20 s per bed or a reduction in administered activity to approximately 22% of current clinical practice, while maintaining spatial resolution, contrast, quantitative accuracy, and diagnostic image quality.
- 22 January 2022
- Published by Springer Nature in Drug Safety
- Vol. 45 (2) , 137-144
Spontaneous Reporting to Regulatory Authorities of Suspected Adverse Drug Reactions to COVID-19 Vaccines Over Time: The Effect of Publicity
Introduction The UK Medicines and Healthcare products Regulatory Agency (MHRA) has published frequent summaries of spontaneous reports of suspected adverse drug reactions (ADRs) (Yellow Cards) to vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The EudraVigilance database has provided similar data for the European Economic Area. Objective Our objective was to characterize the evolution over time of spontaneous reports of suspected ADRs to coronavirus disease 2019 (COVID-19) vaccines and to observe the effect of a publicized reaction (cerebral venous and sinus thrombosis [CVST]) on reporting rates. Methods We used publicly available data on reports of suspected ADRs and doses of vaccine administered, published by the MHRA, EudraVigilance, and the European Centre for Disease Prevention and Control to calculate reporting rates. Results Approximately 4814 Yellow Card reports (23 fatal) per million doses of ChAdOx1 nCoV-19 (AstraZeneca) and 2890 (13 fatal) per million doses of tozinameran (Pfizer/BioNTech) have been lodged. Between 15 March and 31 October 2021, cumulative European reports of CVST rose from 0 to 443 (183 with thrombocytopenia, 72 fatal) with ChAdOx1 nCoV-19 and from 2 to 315 (9 with thrombocytopenia, 28 fatal) with tozinameran. European cases of retinal vein occlusion and thrombosis rose from 0 to 168 with ChAdOx1 nCoV-19 and from 1 to 220 with tozinameran; four of the ChAdOx1 nCoV-19 cases were associated with thrombocytopenia. Conclusion Reports of fatal adverse reactions to coronavirus vaccines are very rare. Reports of CVST have been made in relation to both vaccines. Most were submitted after the reaction had been publicized. Thrombocytopenia occurred in a minority of cases. Reports linked both vaccines to cases of retinal vein thrombosis, just four cases with thrombocytopenia. This suggests two different mechanisms of thrombosis associated with the vaccines.
- 22 January 2022
- Published by Springer Nature in Drug Safety
- Vol. 45 (2) , 137-144
Spontaneous Reporting to Regulatory Authorities of Suspected Adverse Drug Reactions to COVID-19 Vaccines Over Time: The Effect of Publicity
Introduction The UK Medicines and Healthcare products Regulatory Agency (MHRA) has published frequent summaries of spontaneous reports of suspected adverse drug reactions (ADRs) (Yellow Cards) to vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The EudraVigilance database has provided similar data for the European Economic Area. Objective Our objective was to characterize the evolution over time of spontaneous reports of suspected ADRs to coronavirus disease 2019 (COVID-19) vaccines and to observe the effect of a publicized reaction (cerebral venous and sinus thrombosis [CVST]) on reporting rates. Methods We used publicly available data on reports of suspected ADRs and doses of vaccine administered, published by the MHRA, EudraVigilance, and the European Centre for Disease Prevention and Control to calculate reporting rates. Results Approximately 4814 Yellow Card reports (23 fatal) per million doses of ChAdOx1 nCoV-19 (AstraZeneca) and 2890 (13 fatal) per million doses of tozinameran (Pfizer/BioNTech) have been lodged. Between 15 March and 31 October 2021, cumulative European reports of CVST rose from 0 to 443 (183 with thrombocytopenia, 72 fatal) with ChAdOx1 nCoV-19 and from 2 to 315 (9 with thrombocytopenia, 28 fatal) with tozinameran. European cases of retinal vein occlusion and thrombosis rose from 0 to 168 with ChAdOx1 nCoV-19 and from 1 to 220 with tozinameran; four of the ChAdOx1 nCoV-19 cases were associated with thrombocytopenia. Conclusion Reports of fatal adverse reactions to coronavirus vaccines are very rare. Reports of CVST have been made in relation to both vaccines. Most were submitted after the reaction had been publicized. Thrombocytopenia occurred in a minority of cases. Reports linked both vaccines to cases of retinal vein thrombosis, just four cases with thrombocytopenia. This suggests two different mechanisms of thrombosis associated with the vaccines.
- 1 December 2021
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 301 (3) , 692-699
Artificial Intelligence Algorithm Improves Radiologist Performance in Skeletal Age Assessment: A Prospective Multicenter Randomized Controlled Trial
Use of an artificial intelligence algorithm as a diagnostic aid improved the quality of skeletal age assessment in a prospective multicenter setting. Background Previous studies suggest that use of artificial intelligence (AI) algorithms as diagnostic aids may improve the quality of skeletal age assessment, though these studies lack evidence fro...
- 1 December 2021
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 301 (3) , 692-699
Artificial Intelligence Algorithm Improves Radiologist Performance in Skeletal Age Assessment: A Prospective Multicenter Randomized Controlled Trial
Use of an artificial intelligence algorithm as a diagnostic aid improved the quality of skeletal age assessment in a prospective multicenter setting. Background Previous studies suggest that use of artificial intelligence (AI) algorithms as diagnostic aids may improve the quality of skeletal age assessment, though these studies lack evidence fro...
- Top Cited Papers
- 1 January 2021
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 298 (1) , 180-188
Improving Image Quality and Reducing Radiation Dose for Pediatric CT by Using Deep Learning Reconstruction
CT deep learning reconstruction improved image quality, had better object detection performance and radiologist confidence, and may be used for a greater radiation dose reduction potential than alt... Background CT deep learning reconstruction (DLR) algorithms have been developed to remove image noise. How the DLR affects image quality and radiation dose reduction has yet to be fully investigate...
- 1 January 2021
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 298 (1) , 180-188
Top Cited PapersImproving Image Quality and Reducing Radiation Dose for Pediatric CT by Using Deep Learning Reconstruction
CT deep learning reconstruction improved image quality, had better object detection performance and radiologist confidence, and may be used for a greater radiation dose reduction potential than alt... Background CT deep learning reconstruction (DLR) algorithms have been developed to remove image noise. How the DLR affects image quality and radiation dose reduction has yet to be fully investigate...
- 1 September 2020
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 2 (5) , e190226
Automatic Detection of Inadequate Pediatric Lateral Neck Radiographs of the Airway and Soft Tissues using Deep Learning
Purpose To develop and validate a deep learning (DL) algorithm to identify poor-quality lateral airway radiographs. Materials and Methods A total of 1200 lateral airway radiographs obtained in emergency department patients between January 1, 2000, and July 1, 2019, were retrospectively queried from the picture archiving and communication system. Two radiologists classified each radiograph as adequate or inadequate. Disagreements were adjudicated by a third radiologist. The radiographs were used to train and test the DL classifiers. Three technologists and three different radiologists classified the images in the test dataset, and their performance was compared with that of the DL classifiers. Results The training set had 961 radiographs and the test set had 239. The best DL classifier (ResNet-50) achieved sensitivity, specificity, and area under the receiver operating characteristic curve of 0.90 (95% confidence interval [CI]: 0.86, 0.94), 0.82 (95% CI: 0.76, 0.90), and 0.86 (95% CI: 0.81, 0.91), respectively. Interrater agreement for technologists was fair (Fleiss κ, 0.36 [95% CI: 0.29, 0.43]), while that for radiologists was moderate (Fleiss κ, 0.59 [95% CI: 0.52, 0.66]). Cohen κ value comparing the consensus rating of ResNet-50 iterations from fivefold cross-validation, consensus technologists’ rating, and consensus radiologists’ rating to the ground truth were 0.76 (95% CI: 0.63, 0.89), 0.49 (95% CI: 0.37, 0.61), and 0.66 (95% CI: 0.54, 0.78), respectively. Conclusion The development and validation of DL classifiers to distinguish between adequate and inadequate lateral airway radiographs is reported. The classifiers performed significantly better than a group of technologists and as well as the radiologists. © RSNA, 2020
- 1 September 2020
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 2 (5) , e190226
Automatic Detection of Inadequate Pediatric Lateral Neck Radiographs of the Airway and Soft Tissues using Deep Learning
Purpose To develop and validate a deep learning (DL) algorithm to identify poor-quality lateral airway radiographs. Materials and Methods A total of 1200 lateral airway radiographs obtained in emergency department patients between January 1, 2000, and July 1, 2019, were retrospectively queried from the picture archiving and communication system. Two radiologists classified each radiograph as adequate or inadequate. Disagreements were adjudicated by a third radiologist. The radiographs were used to train and test the DL classifiers. Three technologists and three different radiologists classified the images in the test dataset, and their performance was compared with that of the DL classifiers. Results The training set had 961 radiographs and the test set had 239. The best DL classifier (ResNet-50) achieved sensitivity, specificity, and area under the receiver operating characteristic curve of 0.90 (95% confidence interval [CI]: 0.86, 0.94), 0.82 (95% CI: 0.76, 0.90), and 0.86 (95% CI: 0.81, 0.91), respectively. Interrater agreement for technologists was fair (Fleiss κ, 0.36 [95% CI: 0.29, 0.43]), while that for radiologists was moderate (Fleiss κ, 0.59 [95% CI: 0.52, 0.66]). Cohen κ value comparing the consensus rating of ResNet-50 iterations from fivefold cross-validation, consensus technologists’ rating, and consensus radiologists’ rating to the ground truth were 0.76 (95% CI: 0.63, 0.89), 0.49 (95% CI: 0.37, 0.61), and 0.66 (95% CI: 0.54, 0.78), respectively. Conclusion The development and validation of DL classifiers to distinguish between adequate and inadequate lateral airway radiographs is reported. The classifiers performed significantly better than a group of technologists and as well as the radiologists. © RSNA, 2020
- 17 August 2018
- Published by Springer Nature in Abdominal Radiology
- Vol. 44 (2) , 391-397
Inter-radiologist agreement using Society of Abdominal Radiology-American Gastroenterological Association (SAR-AGA) consensus nomenclature for reporting CT and MR enterography in children and young adults with small bowel Crohn disease
To assess inter-radiologist agreement using the Society of Abdominal Radiology-American Gastroenterological Association (SAR-AGA) consensus recommendations for reporting CT/MR enterography exams in pediatric and young adult small bowel Crohn disease (CD).
- 17 August 2018
- Published by Springer Nature in Abdominal Radiology
- Vol. 44 (2) , 391-397
Inter-radiologist agreement using Society of Abdominal Radiology-American Gastroenterological Association (SAR-AGA) consensus nomenclature for reporting CT and MR enterography in children and young adults with small bowel Crohn disease
To assess inter-radiologist agreement using the Society of Abdominal Radiology-American Gastroenterological Association (SAR-AGA) consensus recommendations for reporting CT/MR enterography exams in pediatric and young adult small bowel Crohn disease (CD).
- 1 June 2012
- Published by Wolters Kluwer Health in Journal of Pediatric Orthopaedics
- Vol. 32 (4) , 378-384
The Intrarater and Interrater Reliability of Glenoid Version and Glenohumeral Subluxation Measurements in Neonatal Brachial Plexus Palsy
Progressive and disabling glenohumeral dysplasia commonly occurs as a secondary deformity in children with neonatal brachial plexus palsy (NBPP). A number of methods for quantifying glenohumeral dysplasia are currently in use; however, the most commonly reported quantitative measures have yet to be validated. The present study assesses the intrarater and interrater reliability of the glenoid version angle (GVA) and percent of the humeral head anterior to the scapular line (PHHA) measurements on axial magnetic resonance images. Axial magnetic resonance images of the shoulder girdle of 25 children with NBPP were selected to represent a wide range of glenohumeral dysplasia severity. An axial image was preselected for each measurement. Six examiners (3 orthopaedic surgeons, 2 musculoskeletal radiologists, and an epidemiologist) digitally measured the GVA and PHHA on each image twice, with each measurement separated by 2 to 14 days and the order of image presentation placed in a different arrangement for each measurement set. Intrarater and interrater reliability was assessed with the intraclass correlation coefficient (ICC). Measurement errors for the GVA and PHHA measurements and the variances associated with the scapular and glenoid lines were calculated. Using the Fleiss criteria, intrarater reliability was excellent, with ICCs averaging 0.909 (95% CI: 0.840, 0.940) for GVA and 0.891 (95% CI: 0.815, 0.921) for PHHA. Interrater reliability was excellent, with ICCs of 0.848 (95% CI: 0.788, 0.909) for GVA and 0.874 (95% CI: 0.815, 0.934) for PHHA. The GVA and PHHA measurement errors were ±6.4 degrees and ±7.2%, respectively. In a subset of 141 images measured, the between-image variance in the scapular line was greater than the glenoid line by a 1.61:1 ratio. : The present study demonstrates excellent intrarater and interrater reliability of standard measurements of glenohumeral dysplasia in NBPP. The measurement errors for both measurements were comparable with other standard measures (e.g., Cobb angle). The scapular line exhibited a greater variance than the glenoid line, which identifies an opportunity for improvement in the GVA measurement. Diagnostic study; level III.
- 1 June 2012
- Published by Wolters Kluwer Health in Journal of Pediatric Orthopaedics
- Vol. 32 (4) , 378-384
The Intrarater and Interrater Reliability of Glenoid Version and Glenohumeral Subluxation Measurements in Neonatal Brachial Plexus Palsy
Progressive and disabling glenohumeral dysplasia commonly occurs as a secondary deformity in children with neonatal brachial plexus palsy (NBPP). A number of methods for quantifying glenohumeral dysplasia are currently in use; however, the most commonly reported quantitative measures have yet to be validated. The present study assesses the intrarater and interrater reliability of the glenoid version angle (GVA) and percent of the humeral head anterior to the scapular line (PHHA) measurements on axial magnetic resonance images. Axial magnetic resonance images of the shoulder girdle of 25 children with NBPP were selected to represent a wide range of glenohumeral dysplasia severity. An axial image was preselected for each measurement. Six examiners (3 orthopaedic surgeons, 2 musculoskeletal radiologists, and an epidemiologist) digitally measured the GVA and PHHA on each image twice, with each measurement separated by 2 to 14 days and the order of image presentation placed in a different arrangement for each measurement set. Intrarater and interrater reliability was assessed with the intraclass correlation coefficient (ICC). Measurement errors for the GVA and PHHA measurements and the variances associated with the scapular and glenoid lines were calculated. Using the Fleiss criteria, intrarater reliability was excellent, with ICCs averaging 0.909 (95% CI: 0.840, 0.940) for GVA and 0.891 (95% CI: 0.815, 0.921) for PHHA. Interrater reliability was excellent, with ICCs of 0.848 (95% CI: 0.788, 0.909) for GVA and 0.874 (95% CI: 0.815, 0.934) for PHHA. The GVA and PHHA measurement errors were ±6.4 degrees and ±7.2%, respectively. In a subset of 141 images measured, the between-image variance in the scapular line was greater than the glenoid line by a 1.61:1 ratio. : The present study demonstrates excellent intrarater and interrater reliability of standard measurements of glenohumeral dysplasia in NBPP. The measurement errors for both measurements were comparable with other standard measures (e.g., Cobb angle). The scapular line exhibited a greater variance than the glenoid line, which identifies an opportunity for improvement in the GVA measurement. Diagnostic study; level III.