Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network
Abstract Background Gastroenteropancreatic neuroendocrine tumors most commonly metastasize to the liver; however, high normal background 68Ga-DOTATATE activity and high image noise make metastatic lesions difficult to detect. The purpose of this study is to develop a rapid, automated and highly spec...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2021-10-01
|
| Series: | EJNMMI Research |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13550-021-00839-x |
| _version_ | 1818691482904690688 |
|---|---|
| author | Jonathan Wehrend Michael Silosky Fuyong Xing Bennett B. Chin |
| author_facet | Jonathan Wehrend Michael Silosky Fuyong Xing Bennett B. Chin |
| author_sort | Jonathan Wehrend |
| collection | DOAJ |
| description | Abstract Background Gastroenteropancreatic neuroendocrine tumors most commonly metastasize to the liver; however, high normal background 68Ga-DOTATATE activity and high image noise make metastatic lesions difficult to detect. The purpose of this study is to develop a rapid, automated and highly specific method to identify 68Ga-DOTATATE PET/CT hepatic lesions using a 2D U-Net convolutional neural network. Methods A retrospective study of 68Ga-DOTATATE PET/CT patient studies (n = 125; 57 with 68Ga-DOTATATE hepatic lesions and 68 without) was evaluated. The dataset was randomly divided into 75 studies for the training set (36 abnormal, 39 normal), 25 for the validation set (11 abnormal, 14 normal) and 25 for the testing set (11 abnormal, 14 normal). Hepatic lesions were physician annotated using a modified PERCIST threshold, and boundary definition by gradient edge detection. The 2D U-Net was trained independently five times for 100,000 iterations using a linear combination of binary cross-entropy and dice losses with a stochastic gradient descent algorithm. Performance metrics included: positive predictive value (PPV), sensitivity, F1 score and area under the precision–recall curve (PR-AUC). Five different pixel area thresholds were used to filter noisy predictions. Results A total of 233 lesions were annotated with each abnormal study containing a mean of 4 ± 2.75 lesions. A pixel filter of 20 produced the highest mean PPV 0.94 ± 0.01. A pixel filter of 5 produced the highest mean sensitivity 0.74 ± 0.02. The highest mean F1 score 0.79 ± 0.01 was produced with a 20 pixel filter. The highest mean PR-AUC 0.73 ± 0.03 was produced with a 15 pixel filter. Conclusion Deep neural networks can automatically detect hepatic lesions in 68Ga-DOTATATE PET. Ongoing improvements in data annotation methods, increasing sample sizes and training methods are anticipated to further improve detection performance. |
| first_indexed | 2024-12-17T12:42:36Z |
| format | Article |
| id | doaj.art-760fbec994584216ac52aaa4b5891888 |
| institution | Directory Open Access Journal |
| issn | 2191-219X |
| language | English |
| last_indexed | 2024-12-17T12:42:36Z |
| publishDate | 2021-10-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EJNMMI Research |
| spelling | doaj.art-760fbec994584216ac52aaa4b58918882022-12-21T21:47:55ZengSpringerOpenEJNMMI Research2191-219X2021-10-0111111110.1186/s13550-021-00839-xAutomated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural networkJonathan Wehrend0Michael Silosky1Fuyong Xing2Bennett B. Chin3Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, University of Colorado School of Medicine Anschutz Medical CampusDivision of Nuclear Medicine and Molecular Imaging, Department of Radiology, University of Colorado School of Medicine Anschutz Medical CampusDepartment of Biostatistics and Informatics Colorado School of Public Health, University of Colorado Anschutz Medical CampusDivision of Nuclear Medicine and Molecular Imaging, Department of Radiology, University of Colorado School of Medicine Anschutz Medical CampusAbstract Background Gastroenteropancreatic neuroendocrine tumors most commonly metastasize to the liver; however, high normal background 68Ga-DOTATATE activity and high image noise make metastatic lesions difficult to detect. The purpose of this study is to develop a rapid, automated and highly specific method to identify 68Ga-DOTATATE PET/CT hepatic lesions using a 2D U-Net convolutional neural network. Methods A retrospective study of 68Ga-DOTATATE PET/CT patient studies (n = 125; 57 with 68Ga-DOTATATE hepatic lesions and 68 without) was evaluated. The dataset was randomly divided into 75 studies for the training set (36 abnormal, 39 normal), 25 for the validation set (11 abnormal, 14 normal) and 25 for the testing set (11 abnormal, 14 normal). Hepatic lesions were physician annotated using a modified PERCIST threshold, and boundary definition by gradient edge detection. The 2D U-Net was trained independently five times for 100,000 iterations using a linear combination of binary cross-entropy and dice losses with a stochastic gradient descent algorithm. Performance metrics included: positive predictive value (PPV), sensitivity, F1 score and area under the precision–recall curve (PR-AUC). Five different pixel area thresholds were used to filter noisy predictions. Results A total of 233 lesions were annotated with each abnormal study containing a mean of 4 ± 2.75 lesions. A pixel filter of 20 produced the highest mean PPV 0.94 ± 0.01. A pixel filter of 5 produced the highest mean sensitivity 0.74 ± 0.02. The highest mean F1 score 0.79 ± 0.01 was produced with a 20 pixel filter. The highest mean PR-AUC 0.73 ± 0.03 was produced with a 15 pixel filter. Conclusion Deep neural networks can automatically detect hepatic lesions in 68Ga-DOTATATE PET. Ongoing improvements in data annotation methods, increasing sample sizes and training methods are anticipated to further improve detection performance.https://doi.org/10.1186/s13550-021-00839-xDeep learningConvolutional neural networkNeuroendocrine tumorDOTATATESomatostatin receptorPositron emission tomography |
| spellingShingle | Jonathan Wehrend Michael Silosky Fuyong Xing Bennett B. Chin Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network EJNMMI Research Deep learning Convolutional neural network Neuroendocrine tumor DOTATATE Somatostatin receptor Positron emission tomography |
| title | Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network |
| title_full | Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network |
| title_fullStr | Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network |
| title_full_unstemmed | Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network |
| title_short | Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network |
| title_sort | automated liver lesion detection in 68ga dotatate pet ct using a deep fully convolutional neural network |
| topic | Deep learning Convolutional neural network Neuroendocrine tumor DOTATATE Somatostatin receptor Positron emission tomography |
| url | https://doi.org/10.1186/s13550-021-00839-x |
| work_keys_str_mv | AT jonathanwehrend automatedliverlesiondetectionin68gadotatatepetctusingadeepfullyconvolutionalneuralnetwork AT michaelsilosky automatedliverlesiondetectionin68gadotatatepetctusingadeepfullyconvolutionalneuralnetwork AT fuyongxing automatedliverlesiondetectionin68gadotatatepetctusingadeepfullyconvolutionalneuralnetwork AT bennettbchin automatedliverlesiondetectionin68gadotatatepetctusingadeepfullyconvolutionalneuralnetwork |