George Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
George Zhou has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
Candace Chien
Children's Hospital Los Angeles, Los Angeles, CA 90027, USA
4 shared publicationsYunchan Chen
Weill Cornell Medicine, New York, NY 10021, USA
3 shared publications- Yunchan Chen
Weill Cornell Medical College, New York, NY, USA
2 shared publications Alan Wang
School of Electrical and Computer Engineering, Cornell University and Cornell Tech, New York, USA
1 shared publicationBenjamin Lee
Department of Radiology, Weill Cornell Medicine, New York, USA
1 shared publicationBobak Mosadegh
Dalio Institute of Cardiovascular Imaging, Department of Radiology, Weill Cornell Medicine, New York, New York
1 shared publication
Latest Publications
- 4 April 2024
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 153, 102867
Identifying pediatric heart murmurs and distinguishing innocent from pathologic using deep learning
- 4 April 2024
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 153, 102867
Identifying pediatric heart murmurs and distinguishing innocent from pathologic using deep learning
- Open Access
- 2 September 2023
- Published by Elsevier in Academic Radiology
- Vol. 31 (1) , 104-120
Distilling Knowledge From an Ensemble of Vision Transformers for Improved Classification of Breast Ultrasound
- 2 September 2023
- Published by Elsevier in Academic Radiology
- Vol. 31 (1) , 104-120
Open AccessDistilling Knowledge From an Ensemble of Vision Transformers for Improved Classification of Breast Ultrasound
- Open Access
- 1 September 2023
- Published by Springer Nature in npj Digital Medicine
- Vol. 6 (1) , 1-14
Deep learning analysis of blood flow sounds to detect arteriovenous fistula stenosis
For hemodialysis patients, arteriovenous fistula (AVF) patency determines whether adequate hemofiltration can be achieved, and directly influences clinical outcomes. Here, we report the development and performance of a deep learning model for automated AVF stenosis screening based on the sound of AVF blood flow using supervised learning with data validated by ultrasound. We demonstrate the importance of contextualizing the sound with location metadata as the characteristics of the blood flow sound varies significantly along the AVF. We found the best model to be a vision transformer trained on spectrogram images. Our model can screen for stenosis at a performance level comparable to that of a nephrologist performing a physical exam, but with the advantage of being automated and scalable. In a high-volume, resource-limited clinical setting, automated AVF stenosis screening can help ensure patient safety via early detection of at-risk vascular access, streamline the dialysis workflow, and serve as a patient-facing tool to allow for at-home, self-screening.
- 1 September 2023
- Published by Springer Nature in npj Digital Medicine
- Vol. 6 (1) , 1-14
Open AccessDeep learning analysis of blood flow sounds to detect arteriovenous fistula stenosis
For hemodialysis patients, arteriovenous fistula (AVF) patency determines whether adequate hemofiltration can be achieved, and directly influences clinical outcomes. Here, we report the development and performance of a deep learning model for automated AVF stenosis screening based on the sound of AVF blood flow using supervised learning with data validated by ultrasound. We demonstrate the importance of contextualizing the sound with location metadata as the characteristics of the blood flow sound varies significantly along the AVF. We found the best model to be a vision transformer trained on spectrogram images. Our model can screen for stenosis at a performance level comparable to that of a nephrologist performing a physical exam, but with the advantage of being automated and scalable. In a high-volume, resource-limited clinical setting, automated AVF stenosis screening can help ensure patient safety via early detection of at-risk vascular access, streamline the dialysis workflow, and serve as a patient-facing tool to allow for at-home, self-screening.
- Open Access
- 2 August 2023
- Published by Elsevier in JPRAS Open
- Vol. 38, 1-13
Machine-Learning Prediction of Capsular Contraction after Two-Stage Breast Reconstruction
- 2 August 2023
- Published by Elsevier in JPRAS Open
- Vol. 38, 1-13
Open AccessMachine-Learning Prediction of Capsular Contraction after Two-Stage Breast Reconstruction
- Open Access
- 29 August 2022
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 22 (1) , 1-21
On the analysis of data augmentation methods for spectral imaged based heart sound classification using convolutional neural networks
Background: The application of machine learning to cardiac auscultation has the potential to improve the accuracy and efficiency of both routine and point-of-care screenings. The use of convolutional neural networks (CNN) on heart sound spectrograms in particular has defined state-of-the-art performance. However, the relative paucity of patient data remains a significant barrier to creating models that can adapt to a wide range of potential variability. To that end, we examined a CNN model’s performance on automated heart sound classification, before and after various forms of data augmentation, and aimed to identify the most optimal augmentation methods for cardiac spectrogram analysis. Results: We built a standard CNN model to classify cardiac sound recordings as either normal or abnormal. The baseline control model achieved a PR AUC of 0.763 ± 0.047. Among the single data augmentation techniques explored, horizontal flipping of the spectrogram image improved the model performance the most, with a PR AUC of 0.819 ± 0.044. Principal component analysis color augmentation (PCA) and perturbations of saturation-value (SV) of the hue-saturation-value (HSV) color scale achieved a PR AUC of 0.779 ± 045 and 0.784 ± 0.037, respectively. Time and frequency masking resulted in a PR AUC of 0.772 ± 0.050. Pitch shifting, time stretching and compressing, noise injection, vertical flipping, and applying random color filters negatively impacted model performance. Concatenating the best performing data augmentation technique (horizontal flip) with PCA and SV perturbations improved model performance. Conclusion: Data augmentation can improve classification accuracy by expanding and diversifying the dataset, which protects against overfitting to random variance. However, data augmentation is necessarily domain specific. For example, methods like noise injection have found success in other areas of automated sound classification, but in the context of cardiac sound analysis, noise injection can mimic the presence of murmurs and worsen model performance. Thus, care should be taken to ensure clinically appropriate forms of data augmentation to avoid negatively impacting model performance.
- 29 August 2022
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 22 (1) , 1-21
Open AccessOn the analysis of data augmentation methods for spectral imaged based heart sound classification using convolutional neural networks
Background: The application of machine learning to cardiac auscultation has the potential to improve the accuracy and efficiency of both routine and point-of-care screenings. The use of convolutional neural networks (CNN) on heart sound spectrograms in particular has defined state-of-the-art performance. However, the relative paucity of patient data remains a significant barrier to creating models that can adapt to a wide range of potential variability. To that end, we examined a CNN model’s performance on automated heart sound classification, before and after various forms of data augmentation, and aimed to identify the most optimal augmentation methods for cardiac spectrogram analysis. Results: We built a standard CNN model to classify cardiac sound recordings as either normal or abnormal. The baseline control model achieved a PR AUC of 0.763 ± 0.047. Among the single data augmentation techniques explored, horizontal flipping of the spectrogram image improved the model performance the most, with a PR AUC of 0.819 ± 0.044. Principal component analysis color augmentation (PCA) and perturbations of saturation-value (SV) of the hue-saturation-value (HSV) color scale achieved a PR AUC of 0.779 ± 045 and 0.784 ± 0.037, respectively. Time and frequency masking resulted in a PR AUC of 0.772 ± 0.050. Pitch shifting, time stretching and compressing, noise injection, vertical flipping, and applying random color filters negatively impacted model performance. Concatenating the best performing data augmentation technique (horizontal flip) with PCA and SV perturbations improved model performance. Conclusion: Data augmentation can improve classification accuracy by expanding and diversifying the dataset, which protects against overfitting to random variance. However, data augmentation is necessarily domain specific. For example, methods like noise injection have found success in other areas of automated sound classification, but in the context of cardiac sound analysis, noise injection can mimic the presence of murmurs and worsen model performance. Thus, care should be taken to ensure clinically appropriate forms of data augmentation to avoid negatively impacting model performance.
- Open Access
- 20 September 2021
- Published by Springer Nature in Research Square
On the Analysis of Data Augmentation Methods for Spectral Imaged Based Heart Sound Classification Using Convolutional Neural Networks
Background: The application of machine learning to cardiac auscultation has the potential to improve the accuracy and efficiency of both routine and point-of-care screenings. The use of Convolutional Neural Networks (CNN) on heart sound spectrograms in particular has defined state-of-the-art performance. However, the relative paucity of patient data remains a significant barrier to creating models that can adapt to the wide range of between-subject variability. To that end, we examined a CNN model’s performance on automated heart sound classification, before and after various forms of data augmentation, and aimed to identify the most optimal augmentation methods for cardiac spectrogram analysis.Results: We built a standard CNN model to classify cardiac sound recordings as either normal or abnormal. The baseline control model achieved an ROC AUC of 0.945±0.016. Among the data augmentation techniques explored, horizontal flipping of the spectrogram image improved the model performance the most, with an ROC AUC of 0.957±0.009. Principal component analysis color augmentation (PCA) and perturbations of saturation-value (SV) of the hue-saturation-value (HSV) color scale achieved an ROC AUC of 0.949±0.014 and 0.946±0.019, respectively. Time and frequency masking resulted in an ROC AUC of 0.948±0.012. Pitch shifting, time stretching and compressing, noise injection, vertical flipping, and applying random color filters all negatively impacted model performance.Conclusion: Data augmentation can improve classification accuracy by expanding and diversifying the dataset, which protects against overfitting to random variance. However, data augmentation is necessarily domain specific. For example, methods like noise injection have found success in other areas of automated sound classification, but in the context of cardiac sound analysis, noise injection can mimic the presence of murmurs and worsen model performance. Thus, care should be taken to ensure clinically appropriate forms of data augmentation to avoid negatively impacting model performance.
- 20 September 2021
- Published by Springer Nature in Research Square
Open AccessOn the Analysis of Data Augmentation Methods for Spectral Imaged Based Heart Sound Classification Using Convolutional Neural Networks
Background: The application of machine learning to cardiac auscultation has the potential to improve the accuracy and efficiency of both routine and point-of-care screenings. The use of Convolutional Neural Networks (CNN) on heart sound spectrograms in particular has defined state-of-the-art performance. However, the relative paucity of patient data remains a significant barrier to creating models that can adapt to the wide range of between-subject variability. To that end, we examined a CNN model’s performance on automated heart sound classification, before and after various forms of data augmentation, and aimed to identify the most optimal augmentation methods for cardiac spectrogram analysis.Results: We built a standard CNN model to classify cardiac sound recordings as either normal or abnormal. The baseline control model achieved an ROC AUC of 0.945±0.016. Among the data augmentation techniques explored, horizontal flipping of the spectrogram image improved the model performance the most, with an ROC AUC of 0.957±0.009. Principal component analysis color augmentation (PCA) and perturbations of saturation-value (SV) of the hue-saturation-value (HSV) color scale achieved an ROC AUC of 0.949±0.014 and 0.946±0.019, respectively. Time and frequency masking resulted in an ROC AUC of 0.948±0.012. Pitch shifting, time stretching and compressing, noise injection, vertical flipping, and applying random color filters all negatively impacted model performance.Conclusion: Data augmentation can improve classification accuracy by expanding and diversifying the dataset, which protects against overfitting to random variance. However, data augmentation is necessarily domain specific. For example, methods like noise injection have found success in other areas of automated sound classification, but in the context of cardiac sound analysis, noise injection can mimic the presence of murmurs and worsen model performance. Thus, care should be taken to ensure clinically appropriate forms of data augmentation to avoid negatively impacting model performance.