Abstract
Objectives
To design a deep learning-based framework for automatic segmentation and detection of intracranial aneurysms (IAs) on magnetic resonance T1 images and test the robustness and performance of framework.
Methods
A retrospective diagnostic study was conducted based on 159 IAs from 136 patients who underwent the T1 images. Among them, 127 cases were randomly selected for training and validation, and 32 cases were used to assess the accuracy and consistency of our algorithm. We developed and assembled three convolutional neural networks for the segmentation and detection of IAs. The segmentation and detection performance of the model were compared with the ground truth, and various metrics were calculated at the voxel level, IAs level, and patient level to show the performance of our framework.
Results
Our assembled model achieved overall Dice, voxel-level sensitivity, specificity, balanced accuracy, and F1 score of 0.802, 0.874, 0.9998, 0.937, and 0.802, respectively. A coincidence greater than 0.7 between the aneurysms predicted by the model and the ground truth was considered as a true positive. For IAs detection, the sensitivity reached 90.63% with 0.58 false positives per case. The volume of IAs segmented by our model showed a high agreement and consistency with the volume of IAs labeled by experts.
Conclusion
The deep learning framework is achievable and robust for IAs segmentation and detection. Our model offers more clinical application opportunities compared to digital subtraction angiography (DSA)-based, CTA-based, and MRA-based methods.
Clinical relevance statement
Our deep learning framework effectively detects and segments intracranial aneurysms using clinical routine T1 sequences, showing remarkable effectiveness and offering great potential for improving the detection of latent intracranial aneurysms and enabling early identification.
Key Points
•There is no segmentation method based on clinical routine T1 images. Our study shows that the proper deep learning framework can effectively localize the intracranial aneurysms.
•The T1-based segmentation and detection method is more universal than other angiography-based detection methods, which can potentially reduce missed diagnoses caused by the absence of angiography images.
•The deep learning framework is robust and has the potential to be applied in a clinical setting.
Similar content being viewed by others
Abbreviations
- 3D CNN:
-
Three-dimensional convolutional neural network
- CCC:
-
Concordance correlation coefficient
- CLAHE:
-
Contrast limitation adaptive histogram equalization
- CTA:
-
Computer tomography angiography
- DSA:
-
Digital subtraction angiography
- FOV:
-
Field of view
- FPs:
-
False positives
- IAs:
-
Intracranial aneurysms
- MAE:
-
Mean absolute error
- MBC:
-
Minimum bounding cube
- MRA:
-
Magnetic resonance angiography
- ROIs:
-
Regions of interest
- SAH:
-
Subarachnoid hemorrhage
References
Etminan N, Rinkel GJ (2016) Unruptured intracranial aneurysms: development, rupture and preventive management. Nat Rev Neurol 12:699–713. https://doi.org/10.1038/nrneurol.2016.150
Algra AM, Lindgren A, Vergouwen MDI et al (2019) Procedural clinical complications, case-fatality risks, and risk factors in endovascular and neurosurgical treatment of unruptured intracranial aneurysms: a systematic review and meta-analysis. JAMA Neurol 76:282–293. https://doi.org/10.1001/jamaneurol.2018.4165
Walcott BP, Stapleton CJ, Choudhri O, Patel AB (2016) Flow diversion for the treatment of intracranial aneurysms. JAMA Neurol 73:1002–1008. https://doi.org/10.1001/jamaneurol.2016.0609
Lather HD, Gornik HL, Olin JW et al (2017) Prevalence of intracranial aneurysm in women with fibromuscular dysplasia: a report from the US Registry for Fibromuscular Dysplasia. JAMA Neurol 74:1081–1087. https://doi.org/10.1001/jamaneurol.2017.1333
Frösen J, Tulamo R, Paetau A et al (2012) Saccular intracranial aneurysm: pathology and mechanisms. Acta Neuropathol 123:773–786. https://doi.org/10.1007/s00401-011-0939-3
Nam JS, Jeon SB, Jo JY et al (2019) Perioperative rupture risk of unruptured intracranial aneurysms in cardiovascular surgery. Brain 142:1408–1415. https://doi.org/10.1093/brain/awz058
Liu Q, Zhang Y, Yang J et al (2022) The relationship of morphological-hemodynamic characteristics, inflammation, and remodeling of aneurysm wall in unruptured intracranial aneurysms. Transl Stroke Res 13:88–99. https://doi.org/10.1007/s12975-021-00917-1
Duan H, Huang Y, Liu L, Dai H, Chen Y, Zhou L (2019) Automatic detection on intracranial aneurysm from digital subtraction angiography with cascade convolutional neural networks. Biomed Eng Online 18:1–18. https://doi.org/10.1186/s12938-019-0726-2
Jin H, Geng J, Yin Y et al (2020) Fully automated intracranial aneurysm detection and segmentation from digital subtraction angiography series using an end-to-end spatiotemporal deep neural network. J Neurointerv Surg 12:1023–1027. https://doi.org/10.1136/neurintsurg-2020-015824
Khan H, Sharif M, Bibi N, Muhammad N (2019) A novel algorithm for the detection of cerebral aneurysm using sub-band morphological operation. Eur Phys J Plus 134:34. https://doi.org/10.1140/epjp/i2019-12432-6
Rahmany I, Nemmala MEA, Khlifa N, Megdiche H (2019) Automatic detection of intracranial aneurysm using LBP and Fourier descriptor in angiographic images. Int J Comput Assist Radiol Surg 14:1353–1364. https://doi.org/10.1007/s11548-019-01996-0
Zeng Y, Liu X, Xiao N et al (2020) Automatic diagnosis based on spatial information fusion feature for intracranial aneurysm. IEEE Trans Med Imaging 39:1448–1458. https://doi.org/10.1109/TMI.2019.2951439
Timsit C, Soize S, Benaissa A, Portefaix C, Gauvrit J, Pierot L (2016) Contrast-enhanced and time-of-flight MRA at 3T compared with DSA for the follow-up of intracranial aneurysms treated with the WEB device. AJNR Am J Neuroradiol 37:1684–1689. https://doi.org/10.3174/ajnr.A4791
Ahmed SU, Mocco J, Zhang X et al (2019) MRA versus DSA for the follow-up imaging of intracranial aneurysms treated using endovascular techniques: a meta-analysis. J Neurointerv Surg 11:1009–1014. https://doi.org/10.1136/neurintsurg-2019-014936
Dai X, Huang L, Qian Y et al (2020) Deep learning for automated cerebral aneurysm detection on computed tomography images. Int J Comput Assist Radiol Surg 15:715–723. https://doi.org/10.1007/s11548-020-02121-2
Park A, Chute C, Rajpurkar P et al (2019) Deep learning-assisted diagnosis of cerebral aneurysms using the HeadXNet model. JAMA Netw open 2:e195600. https://doi.org/10.1001/jamanetworkopen.2019.5600
Shahzad R, Pennig L, Goertz L et al (2020) Fully automated detection and segmentation of intracranial aneurysms in subarachnoid hemorrhage on CTA using deep learning. Sci Rep 10:1–12. https://doi.org/10.1038/s41598-020-78384-1
Kamnitsas K, Ledig C, Newcombe VFJ et al (2017) Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med Image Anal 36:61–78. https://doi.org/10.1016/j.media.2016.10.004
Yang J, Xie M, Hu C et al (2020) Deep learning for detecting cerebral aneurysms with CT angiography. Radiology 298:155–163. https://doi.org/10.1148/RADIOL.2020192154
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, pp 770–778. https://doi.org/10.1109/CVPR.2016.90
Meng C, Yang D, Chen D (2021) Cerebral aneurysm image segmentation based on multi-modal convolutional neural network. Comput Methods Programs Biomed 208:106285. https://doi.org/10.1016/j.cmpb.2021.106285
Philipp LR, McCracken DJ, McCracken CE et al (2017) Comparison between CTA and digital subtraction angiography in the diagnosis of ruptured aneurysms. Neurosurgery 80:769–777. https://doi.org/10.1093/neuros/nyw113
Bederson JB, Awad IA, Wiebers DO et al (2000) Recommendations for the management of patients with unruptured intracranial aneurysms: a statement for healthcare professionals from the Stroke Council of the American Heart Association. Circulation 102:2300–2308. https://doi.org/10.1161/01.CIR.102.18.2300
Dammert S, Krings T, Ueffing E et al (2004) Detection of intracranial aneurysms with multislice CT: comparison with conventional angiography. Neuroradiology 46:427–434. https://doi.org/10.1007/s00234-003-1155-1
Nakao T, Hanaoka S, Nomura Y et al (2018) Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography. J Magn Reson Imaging 47:948–953. https://doi.org/10.1002/jmri.25842
Chen G, Wei X, Lei H et al (2020) Automated computer-Assisted detection system for cerebral aneurysms in time-of-flight magnetic resonance angiography using fully convolutional network. Biomed Eng Online 19:1–11. https://doi.org/10.1186/s12938-020-00770-7
Sichtermann T, Faron A, Sijben R, Sijben R, Teichert N, Freiherr J, Wiesmann M (2019) Deep learning – based detection of intracranial aneurysms in 3D TOF-MRA. AJNR Am J Neuroradiol 40:25–32
Ueda D, Yamamoto A, Nishimori M et al (2019) Deep learning for MR angiography: automated detection of cerebral aneurysms. Radiology 290:187–194. https://doi.org/10.1148/radiol.2018180901
Joo B, Ahn SS, Yoon PH et al (2020) A deep learning algorithm may automate intracranial aneurysm detection on MR angiography with high diagnostic performance. Eur Radiol 30:5785–5793. https://doi.org/10.1007/s00330-020-06966-8
Li MH, Cheng YS, Li YD et al (2009) Large-cohort comparison between three-dimensional time-of-flight magnetic resonance and rotational digital subtraction angiographies in intracranial aneurysm detection. Stroke 40:3127–3129. https://doi.org/10.1161/STROKEAHA.109.553800
Shi Z, Hu B, Schoepf UJ et al (2020) Artificial intelligence in the management of intracranial aneurysms: current status and future perspectives. AJNR Am J Neuroradiol 41:373–379. https://doi.org/10.3174/AJNR.A6468
Xiang J, Yu J, Snyder KV, Levy EI, Siddiqui AH, Meng H (2016) Hemodynamic-morphological discriminant models for intracranial aneurysm rupture remain stable with increasing sample size. J Neurointerv Surg 8:104–110. https://doi.org/10.1136/neurintsurg-2014-011477
Malhotra A, Wu X, Forman HP et al (2017) Growth and rupture risk of small unruptured intracranial aneurysms a systematic review. Ann Intern Med 167:26–33. https://doi.org/10.7326/M17-0246
Orr JM, Lopez J, Imburgio MJ, Pelletier BA, Bernard JA, Mittal VA (2020) Adolescents at clinical high risk for psychosis show qualitatively altered patterns of activation during rule learning. NeuroImage Clin 27:102286. https://doi.org/10.1016/j.nicl.2020.102286
Yushkevich PA, Piven J, Hazlett HC, Smith RG, Gee JC, Gerig G (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 31:1116–1128. https://doi.org/10.1016/j.neuroimage.2006.01.015
Stimper V, Bauer S, Ernstorfer R, Scholkopf B, Xian RP (2019) Multidimensional contrast limited adaptive histogram equalization. IEEE Access 7:165437–165447. https://doi.org/10.1109/ACCESS.2019.2952899
Huang G, Liu Z, Pleiss G, Lvd M, Weinberger KQ (2019) Convolutional networks with dense connectivity. IEEE Trans Pattern Anal Mach Intell 44:8704–8716. https://doi.org/10.1109/tpami.2019.2918284
Luo X, Liao W, Chen J et al (2021) Efficient semi-supervised gross target volume of nasopharyngeal carcinoma segmentation via uncertainty rectified pyramid consistency. Medical Image Computing and Computer Assisted Intervention (MICCAI), Visual event, pp 318–329. https://doi.org/10.1007/978-3-030-87196-3_30
Ronneberger O, Fischer P, Brox T (2015) U-Net: convolutional networks for biomedical image segmentation. In: Navab N, Hornegger J, Wells WM, Frangi AF (eds) Medical image computing and computer-assisted intervention – MICCAI 2015. Springer International Publishing, Cham, pp 234–241
Wu Y, He K (2020) Group Normalization Int J Comput Vis 128:742–755. https://doi.org/10.1007/s11263-019-01198-w
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: IEEE Internatinal Conference on Computer Vision (ICCV), Santiago, pp 1026–1034. https://doi.org/10.1109/ICCV.2015.123
Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. In: Proceedings of the 32nd International Conference on Machine Learning (PMLR), Lille, vol 37, pp 448–456. http://proceedings.mlr.press/v37/ioffe15.pdf
Nair V, Hinton GE (2010) Rectified linear units improve restricted Boltzmann machines. In: Proceedings of the 27th international conference on machine learning (ICML), Haifa, pp 807-814. https://www.cs.toronto.edu/~hinton/absps/reluICML.pdf
Dou H, Karimi D, Rollins CK et al (2021) A deep attentive convolutional neural network for automatic cortical plate segmentation in fetal MRI. IEEE Trans Med Imaging 40:1123–1133. https://doi.org/10.1109/TMI.2020.3046579
Kingma DP, Ba JL (2015) Adam: a method for stochastic optimization. In: 3rd International Conference on Learning Representation (ICLR), San Diego. https://doi.org/10.48550/arXiv.1412.6980
Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. J Mach Learn Res 9:249–256
Dice LR (1945) Measures of the amount of ecologic association between species. Ecology 26:297–302. https://doi.org/10.2307/1932409
Fenster A, Chiu B (2005) Evaluation of segmentation algorithms for medical imaging. Conf Proc IEEE Eng Med Biol Soc 7:7186–7189. https://doi.org/10.1109/iembs.2005.1616166
Brodersen KH, Ong CS, Stephan KE, Buhmann JM (2010) The balanced accuracy and its posterior distribution. In: 20th International Conference on Pattern Recognition (ICPR), Istanbul, pp 3121–3124. https://doi.org/10.1109/ICPR.2010.764
Hanley JA, McNeil BJ (1982) The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29–36. https://doi.org/10.1148/radiology.143.1.7063747
Rosenfield GH, Fitzpatrick-Lins K (1986) A coefficient of agreement as a measure of thematic classification accuracy. Photogramm Eng Remote Sens 52:223–227
Reddy AR, Prasad EV, Reddy LSS (2013) Abnormality detection of brain MR image segmentation using iterative conditional mode algorithm. Int J Appl Inf Syst 5:56–65
Chicco D, Jurman G (2020) The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics 21:1. https://doi.org/10.1186/s12864-019-6413-7
Müller D, Hartmann D, Meyer P, Auer F, Soto-Rey I, Kramer F (2022) MISeval: a metric library for medical image segmentation evaluation. Stud Health Technol Inform 294:33–37
Giavarina D (2015) Understanding Bland Altman analysis. Biochem Med (Zagreb) 25:141–151. https://doi.org/10.11613/BM.2015.015
Acknowledgements
This manuscript has a pre-print: http://dx.doi.org/10.2139/ssrn.4174298.
Funding
This work was supported by the National Natural Science Foundation of China (Nos. 62171300, 82171290), Natural Science Foundation of Beijing Municipality (Nos. 7222050, L192013), and Beijing Municipal Administration of Hospital’s Ascent Plan (DFL20190501).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Guarantor
The scientific guarantor of this publication is professor Xu Zhang, PhD, from Capital Medical University (zhangxu@ccmu.edu.cn).
Conflict of interest
The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
Statistics and biometry
None.
Informed consent
The institutional review board waived the requirement to obtain informed patient consent for this retrospective study.
Ethical approval
This study was approved by the Institutional Ethical Committee of Beijing Tiantan Hospital.
Study subjects or cohorts overlap
No.
Methodology
• retrospective
• diagnostic or prognostic study
• performed at one institution
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Junda Qu and Hao Niu contributed equally to this work.
Supplementary information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Qu, J., Niu, H., Li, Y. et al. A deep learning framework for intracranial aneurysms automatic segmentation and detection on magnetic resonance T1 images. Eur Radiol (2023). https://doi.org/10.1007/s00330-023-10295-x
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00330-023-10295-x