Real-world Application of a 3D Deep Learning Model for Detecting and Localizing Cerebral Microbleeds

Overview

Journal Acta Neurochir (Wien)

Specialty Neurosurgery

Date 2024 Sep 26

PMID 39325068

Authors

So Yeon Won

Jun-Ho Kim

Changsoo Woo

Dong-Hyun Kim

Keun Young Park

Eung Yeop Kim

Sun-Young Baek

Hyun Jin Han

Beomseok Sohn

Affiliations

Soon will be listed here.

Abstract

Background: Detection and localization of cerebral microbleeds (CMBs) is crucial for disease diagnosis and treatment planning. However, CMB detection is labor-intensive, time-consuming, and challenging owing to its visual similarity to mimics. This study aimed to validate the performance of a three-dimensional (3D) deep learning model that not only detects CMBs but also identifies their anatomic location in real-world settings.

Methods: A total of 21 patients with 116 CMBs and 12 without CMBs were visited in the neurosurgery outpatient department between January 2023 and October 2023. Three readers, including a board-certified neuroradiologist (reader 1), a resident in radiology (reader 2), and a neurosurgeon (reader 3) independently reviewed SWIs of 33 patients to detect CMBs and categorized their locations into lobar, deep, and infratentorial regions without any AI assistance. After a one-month washout period, the same datasets were redistributed randomly, and readers reviewed them again with the assistance of the 3D deep learning model. A comparison of the diagnostic performance between readers with and without AI assistance was performed.

Results: All readers with an AI assistant (reader 1:0.991 [0.930-0.999], reader 2:0.922 [0.881-0.905], and reader 3:0.966 [0.928-0.984]) tended to have higher sensitivity per lesion than readers only (reader 1:0.905 [0.849-0.942], reader 2:0.621 [0.541-0.694], and reader 3:0.871 [0.759-0.935], p = 0.132, 0.017, and 0.227, respectively). In particular, radiology residents (reader 2) showed a statistically significant increase in sensitivity per lesion when using AI. There was no statistically significant difference in the number of FPs per patient for all readers with AI assistant (reader 1: 0.394 [0.152-1.021], reader 2: 0.727 [0.334-1.582], reader 3: 0.182 [0.077-0.429]) and reader only (reader 1: 0.364 [0.159-0.831], reader 2: 0.576 [0.240-1.382], reader 3: 0.121 [0.038-0.383], p = 0.853, 0.251, and 0.157, respectively). Our model accurately categorized the anatomical location of all CMBs.

Conclusions: Our model demonstrated promising potential for the detection and anatomical localization of CMBs, although further research with a larger and more diverse population is necessary to establish clinical utility in real-world settings.

References

Al-Masni M, Kim W, Kim E, Noh Y, Kim D . Automated detection of cerebral microbleeds in MR images: A two-stage deep learning approach. Neuroimage Clin. 2021; 28:102464. PMC: 7575881. DOI: 10.1016/j.nicl.2020.102464. View

Chao C, Kotsenas A, Broderick D . Cerebral amyloid angiopathy: CT and MR imaging findings. Radiographics. 2006; 26(5):1517-31. DOI: 10.1148/rg.265055090. View

Charidimou A, Boulouis G, Frosch M, Baron J, Pasi M, Albucher J . The Boston criteria version 2.0 for cerebral amyloid angiopathy: a multicentre, retrospective, MRI-neuropathology diagnostic accuracy study. Lancet Neurol. 2022; 21(8):714-725. PMC: 9389452. DOI: 10.1016/S1474-4422(22)00208-3. View

Charidimou A, Kakar P, Fox Z, Werring D . Cerebral microbleeds and recurrent stroke risk: systematic review and meta-analysis of prospective ischemic stroke and transient ischemic attack cohorts. Stroke. 2013; 44(4):995-1001. DOI: 10.1161/STROKEAHA.111.000038. View

Charidimou A, Werring D . Cerebral microbleeds and cognition in cerebrovascular disease: an update. J Neurol Sci. 2012; 322(1-2):50-5. DOI: 10.1016/j.jns.2012.05.052. View

Chen Y, Villanueva-Meyer J, Morrison M, Lupo J . Toward Automatic Detection of Radiation-Induced Cerebral Microbleeds Using a 3D Deep Residual Network. J Digit Imaging. 2018; 32(5):766-772. PMC: 6737152. DOI: 10.1007/s10278-018-0146-z. View

Choi K, Kim J, Kang K, Kim J, Choi S, Lee S . Impact of Microbleeds on Outcome Following Recanalization in Patients With Acute Ischemic Stroke. Stroke. 2018; 50(1):127-134. DOI: 10.1161/STROKEAHA.118.023084. View

Fazekas F, Kleinert R, Roob G, Kleinert G, Kapeller P, Schmidt R . Histopathologic analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol. 1999; 20(4):637-42. PMC: 7056037. View

Greenberg S, Vernooij M, Cordonnier C, Viswanathan A, Salman R, Warach S . Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurol. 2009; 8(2):165-74. PMC: 3414436. DOI: 10.1016/S1474-4422(09)70013-4. View

10.

Haller S, Vernooij M, Kuijer J, Larsson E, Jager H, Barkhof F . Cerebral Microbleeds: Imaging and Clinical Significance. Radiology. 2018; 287(1):11-28. DOI: 10.1148/radiol.2018170803. View

11.

Kuijf H, de Bresser J, Geerlings M, Conijn M, Viergever M, Biessels G . Efficient detection of cerebral microbleeds on 7.0 T MR images using the radial symmetry transform. Neuroimage. 2011; 59(3):2266-73. DOI: 10.1016/j.neuroimage.2011.09.061. View

12.

Lee H, Kim J, Lee S, Jung K, Kim W, Noh Y . Detection of Cerebral Microbleeds in MR Images Using a Single-Stage Triplanar Ensemble Detection Network (TPE-Det). J Magn Reson Imaging. 2022; 58(1):272-283. DOI: 10.1002/jmri.28487. View

13.

Lee S, Kim S, Kim N, Yoon B, Roh J . Cortico-subcortical distribution of microbleeds is different between hypertension and cerebral amyloid angiopathy. J Neurol Sci. 2007; 258(1-2):111-4. DOI: 10.1016/j.jns.2007.03.008. View

14.

Liu S, Utriainen D, Chai C, Chen Y, Wang L, Sethi S . Cerebral microbleed detection using Susceptibility Weighted Imaging and deep learning. Neuroimage. 2019; 198:271-282. DOI: 10.1016/j.neuroimage.2019.05.046. View

15.

Myung M, Lee K, Kim H, Oh J, Lee J, Shin I . Novel Approaches to Detection of Cerebral Microbleeds: Single Deep Learning Model to Achieve a Balanced Performance. J Stroke Cerebrovasc Dis. 2021; 30(9):105886. DOI: 10.1016/j.jstrokecerebrovasdis.2021.105886. View

16.

Nakata-Kudo Y, Mizuno T, Yamada K, Shiga K, Yoshikawa K, Mori S . Microbleeds in Alzheimer disease are more related to cerebral amyloid angiopathy than cerebrovascular disease. Dement Geriatr Cogn Disord. 2006; 22(1):8-14. DOI: 10.1159/000092958. View

17.

Passos J, Nzwalo H, Valente M, Marques J, Azevedo A, Netto E . Microbleeds and cavernomas after radiotherapy for paediatric primary brain tumours. J Neurol Sci. 2016; 372:413-416. DOI: 10.1016/j.jns.2016.11.005. View

18.

Roob G, Schmidt R, Kapeller P, Lechner A, Hartung H, Fazekas F . MRI evidence of past cerebral microbleeds in a healthy elderly population. Neurology. 1999; 52(5):991-4. DOI: 10.1212/wnl.52.5.991. View

19.

Scheid R, Preul C, Gruber O, Wiggins C, von Cramon D . Diffuse axonal injury associated with chronic traumatic brain injury: evidence from T2*-weighted gradient-echo imaging at 3 T. AJNR Am J Neuroradiol. 2003; 24(6):1049-56. PMC: 8149043. View

20.

Shuaib A, Akhtar N, Kamran S, Camicioli R . Management of Cerebral Microbleeds in Clinical Practice. Transl Stroke Res. 2018; 10(5):449-457. DOI: 10.1007/s12975-018-0678-z. View