Yajing Zhang1, Yunyun Duan2, Xiaoyang Wang1, Zhizheng Zhuo2, Sven Haller3,4, Frederik Barkhof5,6, Yaou Liu7. 1. MR Clinical Science, Philips Healthcare, 258 Zhongyuan Rd, Suzhou, SIP, China. 2. Department of Radiology, Beijing Tiantan Hospital, Fengtai District, Capital Medical University, No. 119 the West Southern 4th Ring Road, Beijing, 100070, China. 3. Department of Imaging and Medical Informatics, University of Geneva, Geneva, Switzerland. 4. Faculty of Medicine, University of Geneva, Geneva, Switzerland. 5. Department of Radiology and Nuclear Medicine, Amsterdam University Medical Center, Amsterdam, The Netherlands. 6. Queen Square Institute of Neurology and Center for Medical Image Computing, University College, London, UK. 7. Department of Radiology, Beijing Tiantan Hospital, Fengtai District, Capital Medical University, No. 119 the West Southern 4th Ring Road, Beijing, 100070, China. liuyaou@bjtth.org.
Abstract
PURPOSE: White matter hyperintensity (WMHI) lesions on MR images are an important indication of various types of brain diseases that involve inflammation and blood vessel abnormalities. Automated quantification of the WMHI can be valuable for the clinical management of patients, but existing automated software is often developed for a single type of disease and may not be applicable for clinical scans with thick slices and different scanning protocols. The purpose of the study is to develop and validate an algorithm for automatic quantification of white matter hyperintensity suitable for heterogeneous MRI data with different disease types. METHODS: We developed and evaluated "DeepWML", a deep learning method for fully automated white matter lesion (WML) segmentation of multicentre FLAIR images. We used MRI from 507 patients, including three distinct white matter diseases, obtained in 9 centres, with a wide range of scanners and acquisition protocols. The automated delineation tool was evaluated through quantitative parameters of Dice similarity, sensitivity and precision compared to manual delineation (gold standard). RESULTS: The overall median Dice similarity coefficient was 0.78 (range 0.64 ~ 0.86) across the three disease types and multiple centres. The median sensitivity and precision were 0.84 (range 0.67 ~ 0.94) and 0.81 (range 0.64 ~ 0.92), respectively. The tool's performance increased with larger lesion volumes. CONCLUSION: DeepWML was successfully applied to a wide spectrum of MRI data in the three white matter disease types, which has the potential to improve the practical workflow of white matter lesion delineation.
PURPOSE: White matter hyperintensity (WMHI) lesions on MR images are an important indication of various types of brain diseases that involve inflammation and blood vessel abnormalities. Automated quantification of the WMHI can be valuable for the clinical management of patients, but existing automated software is often developed for a single type of disease and may not be applicable for clinical scans with thick slices and different scanning protocols. The purpose of the study is to develop and validate an algorithm for automatic quantification of white matter hyperintensity suitable for heterogeneous MRI data with different disease types. METHODS: We developed and evaluated "DeepWML", a deep learning method for fully automated white matter lesion (WML) segmentation of multicentre FLAIR images. We used MRI from 507 patients, including three distinct white matter diseases, obtained in 9 centres, with a wide range of scanners and acquisition protocols. The automated delineation tool was evaluated through quantitative parameters of Dice similarity, sensitivity and precision compared to manual delineation (gold standard). RESULTS: The overall median Dice similarity coefficient was 0.78 (range 0.64 ~ 0.86) across the three disease types and multiple centres. The median sensitivity and precision were 0.84 (range 0.67 ~ 0.94) and 0.81 (range 0.64 ~ 0.92), respectively. The tool's performance increased with larger lesion volumes. CONCLUSION: DeepWML was successfully applied to a wide spectrum of MRI data in the three white matter disease types, which has the potential to improve the practical workflow of white matter lesion delineation.
Authors: Mike P Wattjes; Àlex Rovira; David Miller; Tarek A Yousry; Maria P Sormani; Maria P de Stefano; Mar Tintoré; Cristina Auger; Carmen Tur; Massimo Filippi; Maria A Rocca; Franz Fazekas; Ludwig Kappos; Chris Polman Journal: Nat Rev Neurol Date: 2015-09-15 Impact factor: 42.937
Authors: Botian Xu; Yaqiong Chai; Cristina M Galarza; Chau Q Vu; Benita Tamrazi; Bilwaj Gaonkar; Luke Macyszyn; Thomas D Coates; Natasha Lepore; John C Wood Journal: Proc IEEE Int Symp Biomed Imaging Date: 2018-05-24