Automatic Detection of Cerebral Microbleeds From MR Images via 3D Convolutional Neural Networks - PubMed (original) (raw)
. 2016 May;35(5):1182-1195.
doi: 10.1109/TMI.2016.2528129. Epub 2016 Feb 11.
- PMID: 26886975
- DOI: 10.1109/TMI.2016.2528129
Automatic Detection of Cerebral Microbleeds From MR Images via 3D Convolutional Neural Networks
Qi Dou et al. IEEE Trans Med Imaging. 2016 May.
Abstract
Cerebral microbleeds (CMBs) are small haemorrhages nearby blood vessels. They have been recognized as important diagnostic biomarkers for many cerebrovascular diseases and cognitive dysfunctions. In current clinical routine, CMBs are manually labelled by radiologists but this procedure is laborious, time-consuming, and error prone. In this paper, we propose a novel automatic method to detect CMBs from magnetic resonance (MR) images by exploiting the 3D convolutional neural network (CNN). Compared with previous methods that employed either low-level hand-crafted descriptors or 2D CNNs, our method can take full advantage of spatial contextual information in MR volumes to extract more representative high-level features for CMBs, and hence achieve a much better detection accuracy. To further improve the detection performance while reducing the computational cost, we propose a cascaded framework under 3D CNNs for the task of CMB detection. We first exploit a 3D fully convolutional network (FCN) strategy to retrieve the candidates with high probabilities of being CMBs, and then apply a well-trained 3D CNN discrimination model to distinguish CMBs from hard mimics. Compared with traditional sliding window strategy, the proposed 3D FCN strategy can remove massive redundant computations and dramatically speed up the detection process. We constructed a large dataset with 320 volumetric MR scans and performed extensive experiments to validate the proposed method, which achieved a high sensitivity of 93.16% with an average number of 2.74 false positives per subject, outperforming previous methods using low-level descriptors or 2D CNNs by a significant margin. The proposed method, in principle, can be adapted to other biomarker detection tasks from volumetric medical data.
Similar articles
- Automated detection of cerebral microbleeds in MR images: A two-stage deep learning approach.
Al-Masni MA, Kim WR, Kim EY, Noh Y, Kim DH. Al-Masni MA, et al. Neuroimage Clin. 2020;28:102464. doi: 10.1016/j.nicl.2020.102464. Epub 2020 Oct 13. Neuroimage Clin. 2020. PMID: 33395960 Free PMC article. - Single-view 2D CNNs with fully automatic non-nodule categorization for false positive reduction in pulmonary nodule detection.
Eun H, Kim D, Jung C, Kim C. Eun H, et al. Comput Methods Programs Biomed. 2018 Oct;165:215-224. doi: 10.1016/j.cmpb.2018.08.012. Epub 2018 Aug 31. Comput Methods Programs Biomed. 2018. PMID: 30337076 - Automated detection of cerebral microbleeds in patients with Traumatic Brain Injury.
van den Heuvel TL, van der Eerden AW, Manniesing R, Ghafoorian M, Tan T, Andriessen TM, Vande Vyvere T, van den Hauwe L, Ter Haar Romeny BM, Goraj BM, Platel B. van den Heuvel TL, et al. Neuroimage Clin. 2016 Jul 2;12:241-51. doi: 10.1016/j.nicl.2016.07.002. eCollection 2016. Neuroimage Clin. 2016. PMID: 27489772 Free PMC article. - VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images.
Chen H, Dou Q, Yu L, Qin J, Heng PA. Chen H, et al. Neuroimage. 2018 Apr 15;170:446-455. doi: 10.1016/j.neuroimage.2017.04.041. Epub 2017 Apr 23. Neuroimage. 2018. PMID: 28445774 Review. - Cerebral Microbleeds: Imaging and Clinical Significance.
Haller S, Vernooij MW, Kuijer JPA, Larsson EM, Jäger HR, Barkhof F. Haller S, et al. Radiology. 2018 Apr;287(1):11-28. doi: 10.1148/radiol.2018170803. Radiology. 2018. PMID: 29558307 Review.
Cited by
- Towards Accurate Microstructure Estimation via 3D Hybrid Graph Transformer.
Yang J, Jiang H, Tassew T, Sun P, Ma J, Xia Y, Yap PT, Chen G. Yang J, et al. Med Image Comput Comput Assist Interv. 2023 Oct;14227:25-34. doi: 10.1007/978-3-031-43993-3_3. Epub 2023 Oct 1. Med Image Comput Comput Assist Interv. 2023. PMID: 39219989 Free PMC article. - Recent advances in artificial intelligent strategies for tissue engineering and regenerative medicine.
Gharibshahian M, Torkashvand M, Bavisi M, Aldaghi N, Alizadeh A. Gharibshahian M, et al. Skin Res Technol. 2024 Sep;30(9):e70016. doi: 10.1111/srt.70016. Skin Res Technol. 2024. PMID: 39189880 Free PMC article. Review. - Artificial Intelligence, the Digital Surgeon: Unravelling Its Emerging Footprint in Healthcare - The Narrative Review.
Shang Z, Chauhan V, Devi K, Patil S. Shang Z, et al. J Multidiscip Healthc. 2024 Aug 15;17:4011-4022. doi: 10.2147/JMDH.S482757. eCollection 2024. J Multidiscip Healthc. 2024. PMID: 39165254 Free PMC article. Review. - DEEP-LEARNING STRATEGY FOR PULMONARY ARTERY-VEIN CLASSIFICATION OF NON-CONTRAST CT IMAGES.
Nardelli P, Jimenez-Carretero D, Bermejo-Peláez D, Ledesma-Carbayo MJ, Rahaghi FN, San José Estépar R. Nardelli P, et al. Proc IEEE Int Symp Biomed Imaging. 2017 Apr;2017:384-387. doi: 10.1109/isbi.2017.7950543. Epub 2017 Jun 19. Proc IEEE Int Symp Biomed Imaging. 2017. PMID: 39070604 Free PMC article. - Artificial Intelligence-Powered Mammography: Navigating the Landscape of Deep Learning for Breast Cancer Detection.
Al Muhaisen S, Safi O, Ulayan A, Aljawamis S, Fakhoury M, Baydoun H, Abuquteish D. Al Muhaisen S, et al. Cureus. 2024 Mar 26;16(3):e56945. doi: 10.7759/cureus.56945. eCollection 2024 Mar. Cureus. 2024. PMID: 38665752 Free PMC article. Review.
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources