Publikation
Deep learning-based mediastinal lymph node assessment on PET/CT images without pixel-level annotations
Sofija Engelson; Yannic Elser; Malte Maria Sieren; Jan Ehrhardt; Julia Andresen; Stefanie Schierholz; Tobias Keck; Daniel Drömann; Jörg Barkhausen; Heinz Handels
In: Journal of Medical Imaging (JMI), Vol. 13, No. 1, Pages 1-18, SPIE, 2/2026.
Zusammenfassung
Purpose
N-staging, a critical component in cancer diagnostics, quantifies metastatic involvement of lymph nodes and plays an important role in guiding treatment decisions. Manual assessment of lymph nodes on PET/CT scans is time-consuming due to minimal contrast to surrounding tissue and strong heterogeneity of the lymph node’s morphology. To streamline the N-staging process, we propose a deep learning-based algorithm that localizes lymph node stations through atlas-to-patient registration, classifies mediastinal lymph node stations as malignant or benign, and subsequently performs automated N-staging. Notably, our model is trained without any pixel-level annotations, i.e., using image-level classification labels only.
Approach
To address the challenge of training without annotations at the pixel level, we use prior knowledge of the lymph node station locations through atlas-to-patient registration and deduce pseudo-labels for lymph node station groups from the N-stage to enable weakly supervised network training.
Results
The proposed algorithm achieves an accuracy of 0.88±0.02, a sensitivity of 0.72±0.08, and a specificity of 0.90±0.03 for lymph node station classification, which is significantly better than the performance of the standard threshold-based approach used for lymph node assessment in radiological images and an algorithm for PET lesion segmentation that was trained with segmentation masks. For automatic N-staging, the accuracy of 0.63±0.04 is on par with an algorithm that was trained with segmentation masks.
Conclusions
The division of the problem setting into subtasks as well as the integration of prior knowledge enables better or comparable performance of models trained with and without segmentation masks.