Deep learning based detection of enlarged perivascular spaces on brain MRI
Deep learning has been demonstrated effective in many neuroimaging applications. However, in many scenarios, the number of imaging sequences capturing information related to small vessel disease lesions is insufficient to support data-driven techniques. Additionally, cohort-based studies may not alw...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Neuroimage: Reports |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666956023000077 |
| _version_ | 1797870524433432576 |
|---|---|
| author | Tanweer Rashid Hangfan Liu Jeffrey B. Ware Karl Li Jose Rafael Romero Elyas Fadaee Ilya M. Nasrallah Saima Hilal R. Nick Bryan Timothy M. Hughes Christos Davatzikos Lenore Launer Sudha Seshadri Susan R. Heckbert Mohamad Habes |
| author_facet | Tanweer Rashid Hangfan Liu Jeffrey B. Ware Karl Li Jose Rafael Romero Elyas Fadaee Ilya M. Nasrallah Saima Hilal R. Nick Bryan Timothy M. Hughes Christos Davatzikos Lenore Launer Sudha Seshadri Susan R. Heckbert Mohamad Habes |
| author_sort | Tanweer Rashid |
| collection | DOAJ |
| description | Deep learning has been demonstrated effective in many neuroimaging applications. However, in many scenarios, the number of imaging sequences capturing information related to small vessel disease lesions is insufficient to support data-driven techniques. Additionally, cohort-based studies may not always have the optimal or essential imaging sequences for accurate lesion detection. Therefore, it is necessary to determine which imaging sequences are crucial for precise detection. This study introduces a deep learning framework to detect enlarged perivascular spaces (ePVS) and aims to find the optimal combination of MRI sequences for deep learning-based quantification. We implemented an effective lightweight U-Net adapted for ePVS detection and comprehensively investigated different combinations of information from SWI, FLAIR, T1-weighted (T1w), and T2-weighted (T2w) MRI sequences. The experimental results showed that T2w MRI is the most important for accurate ePVS detection, and the incorporation of SWI, FLAIR and T1w MRI in the deep neural network had minor improvements in accuracy and resulted in the highest sensitivity and precision (sensitivity = 0.82, precision = 0.83). The proposed method achieved comparable accuracy at a minimal time cost compared to manual reading. The proposed automated pipeline enables robust and time-efficient readings of ePVS from MR scans and demonstrates the importance of T2w MRI for ePVS detection and the potential benefits of using multimodal images. Furthermore, the model provides whole-brain maps of ePVS, enabling a better understanding of their clinical correlates compared to the clinical rating methods within only a couple of brain regions. |
| first_indexed | 2024-04-10T00:29:50Z |
| format | Article |
| id | doaj.art-6d7e694cb4a14c5598006d5df77dee0b |
| institution | Directory Open Access Journal |
| issn | 2666-9560 |
| language | English |
| last_indexed | 2024-04-10T00:29:50Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Neuroimage: Reports |
| spelling | doaj.art-6d7e694cb4a14c5598006d5df77dee0b2023-03-15T04:29:13ZengElsevierNeuroimage: Reports2666-95602023-03-0131100162Deep learning based detection of enlarged perivascular spaces on brain MRITanweer Rashid0Hangfan Liu1Jeffrey B. Ware2Karl Li3Jose Rafael Romero4Elyas Fadaee5Ilya M. Nasrallah6Saima Hilal7R. Nick Bryan8Timothy M. Hughes9Christos Davatzikos10Lenore Launer11Sudha Seshadri12Susan R. Heckbert13Mohamad Habes14Neuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA; Center for Biomedical Image Computing and Analytics, University of Pennsylvania, Philadelphia, PA, USANeuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA; Center for Biomedical Image Computing and Analytics, University of Pennsylvania, Philadelphia, PA, USADepartment of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, USANeuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USADepartment of Neurology, School of Medicine, Boston University, Boston, MA, USANeuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USACenter for Biomedical Image Computing and Analytics, University of Pennsylvania, Philadelphia, PA, USA; Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, USASaw Swee Hock School of Public Health, National University of Singapore and National University Health System, SingaporeDepartment of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, USA; Department of Diagnostic Medicine, Dell Medical School, University of Texas at Austin, Austin, TX, USADepartment of Internal Medicine and Department of Epidemiology and Prevention, Wake Forest School of Medicine, Winston-Salem, NC, USADepartment of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, USALaboratory of Epidemiology and Population Sciences, National Institute on Aging, National Institutes of Health, Bethesda, MD, USANeuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USADepartment of Epidemiology and Cardiovascular Health Research Unit, University of Washington, Seattle, WA, USANeuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA; Center for Biomedical Image Computing and Analytics, University of Pennsylvania, Philadelphia, PA, USA; Corresponding author. Neuroimage Analytics Laboratory and Biggs Institute Neuroimaging Core, Glenn Biggs Institute for Neurodegenerative Disorders, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA.Deep learning has been demonstrated effective in many neuroimaging applications. However, in many scenarios, the number of imaging sequences capturing information related to small vessel disease lesions is insufficient to support data-driven techniques. Additionally, cohort-based studies may not always have the optimal or essential imaging sequences for accurate lesion detection. Therefore, it is necessary to determine which imaging sequences are crucial for precise detection. This study introduces a deep learning framework to detect enlarged perivascular spaces (ePVS) and aims to find the optimal combination of MRI sequences for deep learning-based quantification. We implemented an effective lightweight U-Net adapted for ePVS detection and comprehensively investigated different combinations of information from SWI, FLAIR, T1-weighted (T1w), and T2-weighted (T2w) MRI sequences. The experimental results showed that T2w MRI is the most important for accurate ePVS detection, and the incorporation of SWI, FLAIR and T1w MRI in the deep neural network had minor improvements in accuracy and resulted in the highest sensitivity and precision (sensitivity = 0.82, precision = 0.83). The proposed method achieved comparable accuracy at a minimal time cost compared to manual reading. The proposed automated pipeline enables robust and time-efficient readings of ePVS from MR scans and demonstrates the importance of T2w MRI for ePVS detection and the potential benefits of using multimodal images. Furthermore, the model provides whole-brain maps of ePVS, enabling a better understanding of their clinical correlates compared to the clinical rating methods within only a couple of brain regions.http://www.sciencedirect.com/science/article/pii/S2666956023000077MRIDeep learningEnlarged perivascular space |
| spellingShingle | Tanweer Rashid Hangfan Liu Jeffrey B. Ware Karl Li Jose Rafael Romero Elyas Fadaee Ilya M. Nasrallah Saima Hilal R. Nick Bryan Timothy M. Hughes Christos Davatzikos Lenore Launer Sudha Seshadri Susan R. Heckbert Mohamad Habes Deep learning based detection of enlarged perivascular spaces on brain MRI Neuroimage: Reports MRI Deep learning Enlarged perivascular space |
| title | Deep learning based detection of enlarged perivascular spaces on brain MRI |
| title_full | Deep learning based detection of enlarged perivascular spaces on brain MRI |
| title_fullStr | Deep learning based detection of enlarged perivascular spaces on brain MRI |
| title_full_unstemmed | Deep learning based detection of enlarged perivascular spaces on brain MRI |
| title_short | Deep learning based detection of enlarged perivascular spaces on brain MRI |
| title_sort | deep learning based detection of enlarged perivascular spaces on brain mri |
| topic | MRI Deep learning Enlarged perivascular space |
| url | http://www.sciencedirect.com/science/article/pii/S2666956023000077 |
| work_keys_str_mv | AT tanweerrashid deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT hangfanliu deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT jeffreybware deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT karlli deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT joserafaelromero deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT elyasfadaee deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT ilyamnasrallah deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT saimahilal deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT rnickbryan deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT timothymhughes deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT christosdavatzikos deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT lenorelauner deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT sudhaseshadri deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT susanrheckbert deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri AT mohamadhabes deeplearningbaseddetectionofenlargedperivascularspacesonbrainmri |