Fig. 1: Overall flowchart and workflow of this study. | npj Precision Oncology

Fig. 1: Overall flowchart and workflow of this study.

From: A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs

Fig. 1

a Flowchart of study population selection showing the inclusion and exclusion criteria for six hospitals. b Work pipeline of this study illustrating the construction of the deep-learning model based on three input MRI channels using internal set data. c After training, we used external dataset to compare the ability of the NAFNet-classifier and the ResNet50-classifier to predict adverse pathology (top panel); we also evaluated the predictive ability of the integrated DL-nomogram for adverse pathology (middle panel) and for BCR-free survival (bottom panel). Abbreviations: RP radical prostatectomy, mpMRI multiparametric magnetic resonance imaging, T2WI T2-weighted magnetic resonance imaging, DWI diffusion-weighted imaging, ADC apparent diffusion coefficient, AP adverse pathology, DL deep learning, BCR biochemical recurrence.

Back to article page