Intracranial aneurysms have a high prevalence in human population. It also has a heavy burden of disease and high mortality rate in the case of rupture. Convolutional neural network(CNN) is a type of deep learning architecture which has been proven powerful to detect intracranial aneurysms.
Methods
Four databases were searched using artificial intelligence, intracranial aneurysms, and synonyms to find eligible studies. Articles which had applied CNN for detection of intracranial aneurisms were included in this review. Sensitivity and specificity of the models and human readers regarding modality, size, and location of aneurysms were sought to be extracted. Random model was the preferred model for analyses using CMA 2 to determine pooled sensitivity and specificity.
Results
Overall, 20 studies were used in this review. Deep learning models could detect intracranial aneurysms with a sensitivity of 90/6% (CI: 87/2–93/2%) and specificity of 94/6% (CI: 0/914–0/966). CTA was the most sensitive modality (92.0%(CI:85/2–95/8%)). Overall sensitivity of the models for aneurysms more than 3 mm was above 98% (98%-100%) and 74.6 for aneurysms less than 3 mm. With the aid of AI, the clinicians’ sensitivity increased to 12/8% and interrater agreement to 0/193.
Conclusion
CNN models had an acceptable sensitivity for detection of intracranial aneurysms, surpassing human readers in some fields. The logical approach for application of deep learning models would be its use as a highly capable assistant. In essence, deep learning models are a groundbreaking technology that can assist clinicians and allow them to diagnose intracranial aneurysms more accurately.
VlakMHAlgraABrandenburgR,et al.Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis. The Lancet Neurology2011; 10: 626–636.
2.
Van GijnJKerrRSRinkelGJ. Subarachnoid haemorrhage. The Lancet2007; 369: 306–318.
3.
InvestigatorsUJ. The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med2012; 366: 2474–2482.
4.
IngallTAsplundKMähönenM,et al.A multinational comparison of subarachnoid hemorrhage epidemiology in the WHO MONICA stroke study. Stroke2000; 31: 1054–1061.
5.
HopJWRinkelGAlgraA,et al.Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke1997; 28: 660–664.
6.
VernooijMWIkramMATangheHL, et al.Incidental findings on brain MRI in the general population. N Engl J Med2007; 357: 1821–1828.
7.
Rajabzadeh-OghazHVarbleNShallwaniH, et al.Computer-Assisted three-dimensional morphology evaluation of intracranial aneurysms. World Neurosurg2018; 119: e541–ee50.
8.
VarbleNRajabzadeh-OghazHWangJ,et al.Differences in morphologic and hemodynamic characteristics for “PHASES-based” intracranial aneurysm locations. AJNR American Journal of Neuroradiology2017; 38: 2105–2110.
9.
Rajabzadeh-OghazHVarbleNDaviesJM, et al.Computer-Assisted adjuncts for aneurysmal morphologic assessment: toward more precise and accurate approaches. Proc SPIE Int Soc Opt Eng2017; 10134: 10134.
10.
GrevingJPWermerMJBrownRDJr., et al. Development of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studies. The Lancet Neurology2014; 13: 59–66.
11.
ChalouhiNDumontASRandazzoC, et al.Management of incidentally discovered intracranial vascular abnormalities. Neurosurg Focus2011; 31: E1.
12.
EricksonBJKorfiatisPAkkusZ,et al.Toolkits and libraries for deep learning. J Digit Imaging2017; 30: 400–405.
13.
LarsonDBChenMCLungrenMP,et al.Performance of a deep-learning neural network model in assessing skeletal maturity on pediatric hand radiographs. Radiology2018; 287: 313–322.
14.
KrizhevskyASutskeverIHintonGE. Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst2012; 25: 1097–1105.
15.
LitjensGKooiTBejnordiBE, et al.A survey on deep learning in medical image analysis. Med Image Anal2017; 42: 60–88.
16.
SetioAAACiompiFLitjensG, et al.Pulmonary nodule detection in CT images: false positive reduction using multi-view convolutional networks. IEEE Trans Med Imaging2016; 35: 1160–1169.
17.
ZillyJBuhmannJMMahapatraD. Glaucoma detection using entropy sampling and ensemble learning for automatic optic cup and disc segmentation. Comput Med Imaging Graph2017; 55: 28–41.
18.
MaghdidHSAsaadATGhafoorKZSadiqASMirjaliliSKhanMK, editors. Diagnosing COVID-19 pneumonia from X-ray and CT images using deep learning and transfer learning algorithms. Multimodal image exploitation and learning 2021; 2021: International Society for Optics and Photonics.
19.
SarvamangalaDKulkarniRV. Convolutional neural networks in medical image understanding: a survey. Evol Intell2021; 15: 1–22.
20.
BoZ-HQiaoHTianC, et al.Toward human intervention-free clinical diagnosis of intracranial aneurysm via deep neural network. Patterns2021; 2: 100197.
21.
HaincNMannilMAnagnostakouV, et al.Deep learning based detection of intracranial aneurysms on digital subtraction angiography: a feasibility study. Neuroradiol J2020; 33: 311–317.
22.
JooBChoiHSAhnSS, et al.A deep learning model with high standalone performance for diagnosis of unruptured intracranial aneurysm. Yonsei Med J2021; 62: 1052–1061.
23.
The EndNote Team. Endnote. EndNote X8 ed. Philadelphia, PA: Clarivate; 2013.
24.
BorensteinMHedgesLHigginsJ,et al.Comprehensive meta-analysis version 2. Englewood: Biostat, 2011.
25.
ParkAChuteCRajpurkarP, et al.Deep learning-assisted diagnosis of cerebral aneurysms using the HeadXNet model. JAMA network Open2019; 2: e195600.
26.
ShahzadRPennigLGoertzL, et al.Fully automated detection and segmentation of intracranial aneurysms in subarachnoid hemorrhage on CTA using deep learning. Sci Rep2020; 10: 21799.
27.
BoZHQiaoHTianC, et al.Toward human intervention-free clinical diagnosis of intracranial aneurysm via deep neural network. Patterns2021; 2: 100197.
28.
PennigLHoyerUCIKrauskopfA, et al.Deep learning assistance increases the detection sensitivity of radiologists for secondary intracranial aneurysms in subarachnoid hemorrhage. Neuroradiology2021; 63: 1985–1994.
29.
ShiZMiaoCSchoepfUJ, et al.A clinically applicable deep-learning model for detecting intracranial aneurysm in computed tomography angiography images. Nat Commun2020; 11.
30.
DaiXHuangLQianY, et al.Deep learning for automated cerebral aneurysm detection on computed tomography images. Int J Comput Assist Radiol Surg2020; 15: 715–723.
31.
YangJXieMHuC, et al.Deep learning for detecting cerebral aneurysms with CT angiography. Radiology2021; 298: 155–163.
32.
FaronASichtermannTTeichertN, et al.Performance of a deep-learning neural network to detect intracranial aneurysms from 3D TOF-MRA compared to human readers. Clin Neuroradiol2020; 30: 591–598.
33.
JooBAhnSSYoonPH, et al.A deep learning algorithm may automate intracranial aneurysm detection on MR angiography with high diagnostic performance. Eur Radiol2020; 30: 5785–5793.
34.
SichtermannTFaronASijbenR,et al.Deep learning-based detection of intracranial aneurysms in 3D TOF-MRA. AJNR Am J Neuroradiol2019; 40: 25–32.
35.
HouWMeiSGuiQ, et al.1D CNN-Based intracranial aneurysms detection in 3D TOF-MRA. Complexity2020; 2020.
36.
SohnBParkKYChoiJ, et al.Deep learning-based software improves clinicians’ detection sensitivity of aneurysms on brain TOF-MRA. American Journal of Neuroradiology2021; 42: 1769–1775.
37.
NakaoTHanaokaSNomuraY, et al.Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography. J Magn Reson Imaging2018; 47: 948–953.
38.
UedaDYamamotoANishimoriM, et al.Deep learning for MR angiography: automated detection of cerebral aneurysms. Radiology2019; 290: 187–194.
39.
StemberJNChangPStemberDM, et al.Convolutional neural networks for the detection and measurement of cerebral aneurysms on magnetic resonance angiography. J Digit Imaging2019; 32: 808–815.
40.
LiuXFengJWuZ, et al.Deep neural network-based detection and segmentation of intracranial aneurysms on 3D rotational DSA. Interv Neuroradiol2021; 27: 648–657.
41.
HaincNMannilMAnagnostakouV, et al.Deep learning based detection of intracranial aneurysms on digital subtraction angiography: a feasibility study. Neuroradiology Journal2020; 33: 311–317.
42.
JinHGengJYinY, et al.Fully automated intracranial aneurysm detection and segmentation from digital subtraction angiography series using an end-to-end spatiotemporal deep neural network. J Neurointerv Surg2020; 12: 1023–1027.
43.
ZengYLiuXXiaoN, et al.Automatic diagnosis based on spatial information fusion feature for intracranial aneurysm. IEEE Trans Med Imaging2020; 39: 1448–1458.
44.
HuynhJDRhodesSCHattonJF,et al.Satisfaction of search in periapical radiograph interpretation. J Endod2021; 47: 291–296.
45.
BedersonJBAwadIAWiebersDO, et al.Recommendations for the management of patients with unruptured intracranial aneurysms. Stroke2000; 31: 2742–2750.
46.
GrevingJPRinkelGJBuskensE,et al.Cost-effectiveness of preventive treatment of intracranial aneurysms: new data and uncertainties. Neurology2009; 73: 258–265.
47.
IkawaFMoritaATominariS, et al.Rupture risk of small unruptured cerebral aneurysms. Journal of Neurosurgery JNS2020; 132: 69–78.
48.
LeeG-JEomK-SLeeC,et al.Rupture of very small intracranial aneurysms: incidence and clinical characteristics. J Cerebrovasc Endovasc Neurosurg2015; 17: 217–222.
49.
MalhotraAWuXFormanHP,et al.Management of tiny unruptured intracranial aneurysms: a comparative effectiveness analysis. JAMA Neurol2018; 75: 27–34.
50.
BenderMTWendtHMonarchT, et al.Small aneurysms account for the majority and increasing percentage of aneurysmal subarachnoid hemorrhage: a 25-year, single institution study. Neurosurgery2018; 83: 692–699.
51.
MoritaAKirinoTHashiK, et al.The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med2012; 366: 2474–2482.
52.
WiebersDOWhisnantJPHustonJ3rd, et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet (London, England)2003; 362: 103–110.
53.
GardijanDHeregaTPremužićV, et al.Comparison between stenting and conservative management of posterior circulation perforator aneurysms: systematic review and case series. Neuroradiology2021; 63: 639–651.
54.
BuellTJDingDRaperDMS, et al.Posterior circulation perforator aneurysms: a proposed management algorithm. J Neurointerv Surg2018; 10: 55–59.
55.
TsianakaEAl-ShawishAPotapovA,et al.Clipping versus coiling in posterior circulation intracranial aneurysms: a meta-analysis. Chinese Neurosurgical Journal2019; 5: 16.
56.
SchievinkWI. Intracranial aneurysms. N Engl J Med1997; 336: 28–40.
57.
HowardBMHuRBarrowJW,et al.Comprehensive review of imaging of intracranial aneurysms and angiographically negative subarachnoid hemorrhage. Neurosurgical Focus FOC2019; 47: E20.
58.
UricchioMGuptaSJakowenkoN, et al.Computed tomography angiography versus digital subtraction angiography for postclipping aneurysm obliteration detection. Stroke2019; 50: 381–388.
59.
PradillaGWicksRTHadelsbergU, et al.Accuracy of computed tomography angiography in the diagnosis of intracranial aneurysms. World Neurosurg2013; 80: 845–852.
60.
ChenWXingWHeZ,et al.Accuracy of 320-detector row nonsubtracted and subtracted volume CT angiography in evaluating small cerebral aneurysms. J Neurosurg2017; 127: 725–731.
61.
KemmlingANoelteIGerigkL,et al.A diagnostic pitfall for intracranial aneurysms in time-of-flight MR angiography: small intracranial lipomas. Am J Roentgenol2008; 190: W62–WW7.
62.
VasconcelosRVrtiskaTJFoleyTA, et al.Reducing iodine contrast volume in CT angiography of the abdominal aorta using integrated tube potential selection and weight-based method without compromising image quality. AJR Am J Roentgenol2017; 208: 552–563.
63.
NagashimaHSakamotoHSanoY,et al.[Fundamental study of DSA images using gadolinium contrast agent]. Nihon Hoshasen Gijutsu Gakkai Zasshi2002; 58: 1369–1376.
BjörkmanJFrösenJTähtinenO, et al.Irregular shape identifies ruptured intracranial aneurysm in subarachnoid hemorrhage patients with multiple aneurysms. Stroke2017; 48: 1986–1989.
66.
ChanVLindsayPMcQuigganJ,et al.Declining admission and mortality rates for subarachnoid hemorrhage in Canada between 2004 and 2015. Stroke2019; 50: 181–184.
67.
BedersonJBAwadIAWiebersDO, et al. Recommendations for the management of patients with unruptured intracranial aneurysms: a statement for healthcare professionals from the stroke council of the American heart association. Stroke2000; 31: 2742–2750.
68.
SchatloBFungCStienenMN, et al.Incidence and outcome of aneurysmal subarachnoid hemorrhage: the Swiss study on subarachnoid hemorrhage (Swiss SOS). Stroke2021; 52: 344–347.
69.
LindbohmJVKaprioJJousilahtiP,et al.Risk factors of sudden death from subarachnoid hemorrhage. Stroke2017; 48: 2399–2404.
70.
PetridisAKKampMACorneliusJF, et al.Aneurysmal subarachnoid hemorrhage. Dtsch Arztebl Int2017; 114: 226–236.
71.
DuboshNMBellolioMFRabinsteinAA,et al.Sensitivity of early brain computed tomography to exclude aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. Stroke2016; 47: 750–755.
72.
MarcoliniEHineJ. Approach to the diagnosis and management of subarachnoid hemorrhage. West J Emerg Med2019; 20: 203–211.
73.
ParkS-HYimM-BLeeC-Y,et al.Intracranial fusiform aneurysms: it’s pathogenesis, clinical characteristics and managements. J Korean Neurosurg Soc2008; 44: 116–123.
74.
Faluk M and De Jesus O. Saccular Aneurysm. [Updated 2021 Aug 30]. In: StatPearls[Internet]. Treasure Island (FL): StatPearls Publishing; 2022.
75.
AwadAJMascitelliJRHarounRR,et al.Endovascular management of fusiform aneurysms in the posterior circulation: the era of flow diversion. Neurosurgical Focus FOC2017; 42: E14.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
