Machine learning-based multiparametric magnetic resonance imaging radiomic model for discrimination of pathological subtypes of craniopharyngioma

《Journal of Magnetic Resonance Imaging》

Based on the mpMRI scan, the entire process from data annotation, preprocessing, feature selection to machine learning modeling analysis is completed on the Darwin AI research platform to effectively identify different classifications of craniopharyngeal carcinoma in a noninvasive way.

Application of computer aided diagnosis system based on multi-stage three-dimensional deep convolutional neural network in lung cancer screening

《Chinese Journal of Radiology》

The upgraded AI-CAD system for pulmonary nodules developed by Yizhun AI with improved detection sensitivity and lowered false positive rate drives the realization of "early diagnosis, early treatment” in clinical settings after being widely used in hospitals in China.

Comparison of deep learning convolutional neural network with radiologists in differentiating benign and malignant thyroid nodules on CT images

《ENDOKRYNOLOGIA POLSKA》

Different CNN models and integrated models were used to differentiate benign and malignant thyroid nodules with high accuracy, which was significantly better than the two senior doctors who have 7 and 26 years of experience respectively.

FocalMix: Semi-supervised learning for 3D medical image detection

CVPR2020

Focalmix, a semi-supervised Learning (SSL) framework for 3D medical image detection was launched to solve operation difficulties and high cost of medical image annotation.