Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network

Jonathan Wehrend; Michael Silosky; Fuyong Xing; Bennett B Chin

doi:10.1186/s13550-021-00839-x

Automated liver lesion detection in ⁶⁸Ga DOTATATE PET/CT using a deep fully convolutional neural network

EJNMMI Res. 2021 Oct 2;11(1):98. doi: 10.1186/s13550-021-00839-x.

Authors

Jonathan Wehrend¹, Michael Silosky¹, Fuyong Xing², Bennett B Chin³

Affiliations

¹ Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, University of Colorado School of Medicine Anschutz Medical Campus, 12401 East 17th Avenue, Mail Stop L954A, Aurora, CO, 80045, USA.
² Department of Biostatistics and Informatics Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA.
³ Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, University of Colorado School of Medicine Anschutz Medical Campus, 12401 East 17th Avenue, Mail Stop L954A, Aurora, CO, 80045, USA. bennett.chin@cuanschutz.edu.

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.

Keywords: Convolutional neural network; DOTATATE; Deep learning; Liver tumor; Neuroendocrine tumor; Positron emission tomography; Somatostatin receptor.