Venkata Veerendranadh Chebrolu1, Suresh E. Joel1, Dattesh D. Shanbhag2, Ananda Narasimha Murthy2, Vivek Vaidya2, Patrice Hervo3, Marc-Antoine Labeyrie4, Catherine Oppenheim4, 5, Rakesh Mullick2
1Medical Image Analysis Lab, GE Global Research, Bangalore , Karnataka, India; 2Medical Image Analysis Lab, GE Global Research, Bangalore, Karnataka, India; 3GE Healthcare, Buc, France; 4Departments of Radiology and Neurology, Centre Hospitalier, Sainte-Anne, Paris, France; 5Universit Paris Descartes , Paris, France
The classification of cerebral regions into normal and infarcted tissues based on Diffusion weighted imaging (DWI) characteristics plays a critical role in MRI based stroke patient management. The use of both DWI images and Apparent Diffusion Coefficient (ADC) maps may achieve better classification accuracy than using ADC or DWI alone. In this work, we presented a dual feature (ADC and DWI) based Receiver Operating Characteristic analysis and proposed a novel acute ischemic infarct classification criteria. Results demonstrate better agreement with ground-truth using dual features (specificity-98%; sensitivity-60%) as compared to ADC alone based infarct segmentation (specificity-96%; sensitivity-53%) in 65 subjects.