notesum.ai

Published at November 22

Comparative Analysis of nnUNet and MedNeXt for Head and Neck Tumor Segmentation in MRI-guided Radiotherapy

eess.IV

cs.CV

Released Date: November 22, 2024

Authors: Nikoo Moradi¹, André Ferreira², Behrus Puladi³, Jens Kleesiek², Emad Fatemizadeh¹, Gijs Luijten², Victor Alves⁴, Jan Egger²

Aff.: ¹Department of Electrical Engineering, Sharif University of Technology, Tehran, Iran; ²Institute for AI in Medicine (IKIM), University Medicine Essen, Essen, Germany; ³Department of Oral and Maxillofacial Surgery, University Hospital RWTH Aachen, Aachen, Germany; ⁴Center Algoritmi / LASI, University of Minho, Braga, Portugal

Arxiv: http://arxiv.org/abs/2411.14752v1

Model	GTVp $\text{DSC}_{\text{agg}}$	GTVn $\text{DSC}_{\text{agg}}$	Mean
nnUNet FullRes	0.7772	0.8517	0.8144
nnUNet ResEnc	0.7873	0.8586	0.8230
nnUNet Cascade	0.7847	0.8550	0.8198
nnUNet Cascade + FullRes	0.7846	0.8573	0.8210
nnUNet Cascade + ResEnc	0.7919	0.8633	0.8276
nnUNet FullRes + ResEnc	0.7896	0.8601	0.8249
nnUNet All Ensembled	0.7889	0.8618	0.8254
MedNeXt Small (Kernel 3)	0.8066	0.8710	0.8388
nnUNet + MedNeXt (Average)	0.7931	0.8166	0.8049