A convolutional neural network for total tumor segmentation in [64Cu]Cu-DOTATATE PET/CT of patients with neuroendocrine neoplasms

• Published in: EJNMMI research Vol. 12; no. 1; p. 30
• Main Authors: Carlsen, Esben Andreas, Lindholm, Kristian, Hindsholm, Amalie, Gæde, Mathias, Ladefoged, Claes Nøhr, Loft, Mathias, Johnbeck, Camilla Bardram, Langer, Seppo Wang, Oturai, Peter, Knigge, Ulrich, Kjaer, Andreas, Andersen, Flemming Littrup
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 28.05.2022; Springer Nature B.V; SpringerOpen
• Subjects: [64Cu]Cu-DOTATATE PET; Artificial intelligence; Artificial neural networks; Cardiac Imaging; Image segmentation; Imaging; Medicine; Medicine & Public Health; Neoplasms; Neuroendocrine neoplasms; Nuclear Medicine; Oncology; Original Research; Orthopedics; Performance evaluation; Positron emission; Prognostication; Radiology; Sensitivity; Tomography; Training; Tumor segmentation; Tumors; Neuroendocrine neoplasms; Cu]Cu-DOTATATE PET; Prognostication; [; Tumor segmentation; Artificial intelligence; [64Cu]Cu-DOTATATE PET
• ISSN: 2191-219X; 2191-219X
• DOI: 10.1186/s13550-022-00901-2

Background Segmentation of neuroendocrine neoplasms (NENs) in [ 64 Cu]Cu-DOTATATE positron emission tomography makes it possible to extract quantitative measures useable for prognostication of patients. However, manual tumor segmentation is cumbersome and time-consuming. Therefore, we aimed to implement and test an artificial intelligence (AI) network for tumor segmentation. Patients with gastroenteropancreatic or lung NEN with [ 64 Cu]Cu-DOTATATE PET/CT performed were included in our training ( n  = 117) and test cohort ( n  = 41). Further, 10 patients with no signs of NEN were included as negative controls. Ground truth segmentations were obtained by a standardized semiautomatic method for tumor segmentation by a physician. The nnU-Net framework was used to set up a deep learning U-net architecture. Dice score, sensitivity and precision were used for selection of the final model. AI segmentations were implemented in a clinical imaging viewer where a physician evaluated performance and performed manual adjustments. Results Cross-validation training was used to generate models and an ensemble model. The ensemble model performed best overall with a lesion-wise dice of 0.850 and pixel-wise dice, precision and sensitivity of 0.801, 0.786 and 0.872, respectively. Performance of the ensemble model was acceptable with some degree of manual adjustment in 35/41 (85%) patients. Final tumor segmentation could be obtained from the AI model with manual adjustments in 5 min versus 17 min for ground truth method, p  < 0.01. Conclusion We implemented and validated an AI model that achieved a high similarity with ground truth segmentation and resulted in faster tumor segmentation. With AI, total tumor segmentation may become feasible in the clinical routine.