Fig. 6: Evaluation of the real-world cohort. | npj Digital Medicine

Fig. 6: Evaluation of the real-world cohort.

From: Machine learning model for differentiating xanthogranulomatous cholecystitis and gallbladder cancer in multicenter largescale study

Fig. 6

a Proportions of XGC and GBC cases within the cohort (n = 124). b ROC curve illustrating LIDGAX’s diagnostic performance in differentiating XGC from GBC. c Radar chart quantitatively comparing LIDGAX’s performance metrics. d Confusion matrix of IDGAX in the real-world cohort. e An XGC case classified by LIDGAX with a 77.0% probability, confirmed by pathology. f A GBC case classified by LIDGAX with a 91.0% probability, also confirmed by pathology. XGC Xanthogranulomatous cholecystitis, GBC gallbladder cancer, LIDGAX LightGBM Intelligent Differentiator for XGC and GBC, AUC area under the curve, PPV positive predictive value, NPV negative predictive value, US ultrasound, CT computerized tomography, MRI magnetic resonance imaging.

Back to article page