Fig. 1: Demonstration of the segmentation model’s predictions. | npj Precision Oncology

Fig. 1: Demonstration of the segmentation model’s predictions.

From: End-to-end deep learning for the diagnosis of pelvic and sacral tumors using non-enhanced MRI: a multi-center study

Fig. 1: Demonstration of the segmentation model’s predictions.

A, B T1-w images of a 62-year-old female patient with neurofibroma; C, D T2-w images of a 44-year-old male patient with mucinous papillary ependymoma; E, F DWI images of a 45-year-old female patient with schwannoma; G, H CET1-w images of the same patient as (A) and (B). The second line shows T1-w, T2-w, DWI and CET1-w images (from left to right) with model’s segmentation and radiologist’s segmentation, darker color represents radiologist’s segmentation.

Back to article page