Improving sensitivity through data augmentation with synthetic lymph node metastases for AI-based analysis of PSMA PET-CT images.
Elin TrägårdhJohannes UlénOlof EnqvistLars EdenbrandtMåns LarssonPublished in: Clinical physiology and functional imaging (2024)
Creating synthetic lymph node metastases, as a form of data augmentation, on [18F]PSMA-1007F]PSMA-1007 PETPET-CT-CT images improved the sensitivity of an AI model for detecting suspected lymph node metastases. However, the number of false positive lesions increased somewhat.
Keyphrases
- pet ct
- lymph node
- positron emission tomography
- deep learning
- artificial intelligence
- neoadjuvant chemotherapy
- sentinel lymph node
- image quality
- computed tomography
- big data
- electronic health record
- convolutional neural network
- dual energy
- contrast enhanced
- soft tissue
- pulmonary embolism
- magnetic resonance imaging
- machine learning
- data analysis
- radiation therapy