Training and assessing convolutional neural network performance in automatic vascular segmentation using Ga-68 DOTATATE PET/CT

• Published in: The international journal of cardiovascular imaging 2024-09, Vol.40 (9), p.1847-1861
• Main Authors: Parry, R., Wright, K., Bellinge, J. W., Ebert, M. A., Rowshanfarzad, P., Francis, R. J., Schultz, C. J.
• Format: Article
• Language: English
• Subjects: Adult; Aged; Aorta; Aortic arch; Aortic Diseases - diagnostic imaging; Artificial neural networks; Automation; Calcification; Calcification (ectopic); Calcium; Calcium influx; Cardiac Imaging; Cardiology; Computed tomography; Contours; Coronary vessels; Deep Learning; Female; Heart; Humans; Image Interpretation, Computer-Assisted; Image processing; Imaging; Male; Medicine; Medicine & Public Health; Middle Aged; Neural networks; Neural Networks, Computer; Observer Variation; Organometallic Compounds - administration & dosage; Original Paper; Performance evaluation; Positron emission; Positron emission tomography; Positron Emission Tomography Computed Tomography; Predictive Value of Tests; Radiology; Radiopharmaceuticals - administration & dosage; Reproducibility of Results; Retrospective Studies; Segmentation; Time measurement; Vascular Calcification - diagnostic imaging; Workflow; Young Adult
• ISSN: 1875-8312; 1569-5794; 1875-8312; 1573-0743
• DOI: 10.1007/s10554-024-03171-2

To evaluate a convolutional neural network’s performance (nnU-Net) in the assessment of vascular contours, calcification and PET tracer activity using Ga-68 DOTATATE PET/CT. Patients who underwent Ga-68 DOTATATE PET/CT imaging over a 12-month period for neuroendocrine investigation were included. Manual cardiac and aortic segmentations were performed by an experienced observer. Scans were randomly allocated in ratio 64:16:20 for training, validation and testing of the nnU-Net model. PET tracer uptake and calcium scoring were compared between segmentation methods and different observers. 116 patients (53.5% female) with a median age of 64.5 years (range 23–79) were included. There were strong, positive correlations between all segmentations (mostly r > 0.98). There were no significant differences between manual and AI segmentation of SUV mean for global cardiac (mean ± SD 0.71 ± 0.22 vs. 0.71 ± 0.22; mean diff 0.001 ± 0.008, p  > 0.05), ascending aorta (mean ± SD 0.44 ± 0.14 vs. 0.44 ± 0.14; mean diff 0.002 ± 0.01, p  > 0.05), aortic arch (mean ± SD 0.44 ± 0.10 vs. 0.43 ± 0.10; mean diff 0.008 ± 0.16, p  > 0.05) and descending aorta (mean ± SD  0.05) contours. There was excellent agreement between the majority of manual and AI segmentation measures (r ≥ 0.80) and in all vascular contour calcium scores. Compared with the manual segmentation approach, the CNN required a significantly lower workflow time. AI segmentation of vascular contours using nnU-Net resulted in very similar measures of PET tracer uptake and vascular calcification when compared to an experienced observer and significantly reduced workflow time.