التفاصيل البيبلوغرافية
| العنوان: |
Automated detection of cerebral microbleeds on MR images using knowledge distillation framework |
| المؤلفون: |
Vaanathi Sundaresan, Christoph Arthofer, Giovanna Zamboni, Andrew G. Murchison, Robert A. Dineen, Peter M. Rothwell, Dorothee P. Auer, Chaoyue Wang, Karla L. Miller, Benjamin C. Tendler, Fidel Alfaro-Almagro, Stamatios N. Sotiropoulos, Nikola Sprigg, Ludovica Griffanti, Mark Jenkinson |
| المصدر: |
Frontiers in Neuroinformatics, Vol 17 (2023) |
| بيانات النشر: |
Frontiers Media S.A., 2023. |
| سنة النشر: |
2023 |
| المجموعة: |
LCC:Neurosciences. Biological psychiatry. Neuropsychiatry |
| مصطلحات موضوعية: |
deep learning, knowledge distillation, detection, susceptibility-weighted image (SWI), quantitative susceptibility mapping (QSM), magnetic resonance imaging, Neurosciences. Biological psychiatry. Neuropsychiatry, RC321-571 |
| الوصف: |
IntroductionCerebral microbleeds (CMBs) are associated with white matter damage, and various neurodegenerative and cerebrovascular diseases. CMBs occur as small, circular hypointense lesions on T2*-weighted gradient recalled echo (GRE) and susceptibility-weighted imaging (SWI) images, and hyperintense on quantitative susceptibility mapping (QSM) images due to their paramagnetic nature. Accurate automated detection of CMBs would help to determine quantitative imaging biomarkers (e.g., CMB count) on large datasets. In this work, we propose a fully automated, deep learning-based, 3-step algorithm, using structural and anatomical properties of CMBs from any single input image modality (e.g., GRE/SWI/QSM) for their accurate detections.MethodsIn our method, the first step consists of an initial candidate detection step that detects CMBs with high sensitivity. In the second step, candidate discrimination step is performed using a knowledge distillation framework, with a multi-tasking teacher network that guides the student network to classify CMB and non-CMB instances in an offline manner. Finally, a morphological clean-up step further reduces false positives using anatomical constraints. We used four datasets consisting of different modalities specified above, acquired using various protocols and with a variety of pathological and demographic characteristics.ResultsOn cross-validation within datasets, our method achieved a cluster-wise true positive rate (TPR) of over 90% with an average of 80 % with different modalities. The python implementation of the proposed method is openly available. |
| نوع الوثيقة: |
article |
| وصف الملف: |
electronic resource |
| اللغة: |
English |
| تدمد: |
1662-5196 |
| Relation: |
https://www.frontiersin.org/articles/10.3389/fninf.2023.1204186/full; https://doaj.org/toc/1662-5196 |
| DOI: |
10.3389/fninf.2023.1204186 |
| URL الوصول: |
https://doaj.org/article/ee9b9d7ea4384217946165ef86d8997d |
| رقم الانضمام: |
edsdoj.9b9d7ea4384217946165ef86d8997d |
| قاعدة البيانات: |
Directory of Open Access Journals |
Full Text Finder