Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
401P Persistent bias in head and neck segmentation: Evidence from MedSAM2
0
Zitationen
8
Autoren
2025
Jahr
Abstract
Automated segmentation of organs-at-risk (OARs) is critical for radiotherapy planning. Task-specific models such as nnUNet have demonstrated subgroup-specific biases, while foundation models (e.g., MedSAM2) are expected to generalize across tasks and populations. Whether these models mitigate or reproduce such biases remains unclear.
Ähnliche Arbeiten
Radiative Transfer
1950 · 8.595 Zit.
Practical cone-beam algorithm
1984 · 6.179 Zit.
Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European organization for research and treatment of cancer (EORTC)
1995 · 4.792 Zit.
Tolerance of normal tissue to therapeutic irradiation
1991 · 4.445 Zit.
Clonogenic assay of cells in vitro
2006 · 4.094 Zit.