Guangyu Dan1,2, Cui Feng1,3, Zheng Zhong1,2, Kaibao Sun1, Muge Karaman1,2, Daoyu Hu3, and Xiaohong Joe Zhou1,2,4
1Center for MR Research, University of Illinois at Chicago, Chicago, IL, United States, 2Department of Bioengineering, University of Illinois at Chicago, Chicago, IL, United States, 3Department of Radiology, Tongji Hospital, Wuhan, China, 4Departments of Radiology and Neurosurgery, University of Illinois at Chicago, Chicago, IL, United States
1Center for MR Research, University of Illinois at Chicago, Chicago, IL, United States, 2Department of Bioengineering, University of Illinois at Chicago, Chicago, IL, United States, 3Department of Radiology, Tongji Hospital, Wuhan, China, 4Departments of Radiology and Neurosurgery, University of Illinois at Chicago, Chicago, IL, United States
A model free machine-learning based approach using diffusion signal attenuation
signatures can detect cervical cancerous tissues with a high accuracy. Acquisition
times can be substantially reduced by using this approach without compromising
diagnostic performance.
Figure 1. Pipeline
used in the MODEM machine learning analysis. SMOTE = synthetic minority
oversampling technique.
Figure
2. A set of first-order feature maps from one representative
patient. The color-coded regions indicate the cancerous and normal ROIs on the
diffusion-weighted images with a b-value of 800 s/mm2.
