作者: Nicolas Brieu , Christos G. Gavriel , Ines P. Nearchou , David J. Harrison , Günter Schmidt
DOI: 10.1038/S41598-019-41595-2
关键词: In patient 、 Muscle invasive 、 TNM Staging 、 Tumour budding 、 Machine learning 、 Disease specific 、 Regression tree model 、 Medicine 、 Bladder cancer 、 Stage ii 、 Artificial intelligence
摘要: Tumour budding has been described as an independent prognostic feature in several tumour types. We report for the first time relationship between and survival evaluated patients with muscle invasive bladder cancer. A machine learning-based methodology was applied to accurately quantify buds across immunofluorescence labelled whole slide images from 100 cancer patients. Furthermore, found be correlated TNM (p = 0.00089) pT (p = 0.0078) staging. novel classification regression tree model constructed stratify all stage II, III, IV into three new staging criteria based on disease specific survival. For stratification of non-metastatic high or low risk death, our decision reported that most significant (HR = 2.59, p = 0.0091), no clinical utilised categorise these Our findings demonstrate budding, quantified using automated image analysis provides value a better fit than