Intracranial aneurysm segmentation with nnU-net: utilizing loss functions and automated vessel extraction

Abstract

Aim: Intracranial aneurysms pose significant challenges in diagnosis and treatment, emphasizing the need for accurate segmentation methods to assist clinicians in their management. In this paper, we present a novel approach for segmenting intracranial aneurysms using three-dimensional time-of-flight magnetic resonance angiography (TOF-MRA) images and the no-new-U-net framework. We aim to improve segmentation accuracy and efficiency through the integration of hybrid loss functions and additional vessel information.

Methods: The model was conducted on Aneurysm Detection And SegMentation (ADAM) and Renji Hospital (RENJI) datasets. The TOF-MRA ADAM dataset contains data from 113 cases, where 89 have at least one aneurysm with a median maximum diameter of 3.6 mm and range from 1.0 to 15.9 mm. The RENJI private TOF-MRA dataset comprises 213 cases including both ruptured and unruptured aneurysms with a median maximum diameter of 9.35 mm (range: 1.25-37.58 mm). We optimized the segmentation model by exploring hybrid loss functions that combine distribution-based and region-based losses to effectively delineate intricate aneurysm structures. Additionally, we incorporated vessel information as a region of interest using an automatic vessel segmentation algorithm to enhance the model's focus on critical regions. The model was trained on multi-modality data, including both vessel-enhanced and original images, to capture complementary information and improve segmentation accuracy.

Results: Extensive simulations on both the ADAM dataset and a private RENJI dataset demonstrate the effectiveness of our approach. The best-performing loss function yielded significant improvements in the Dice coefficient (0.72 and 0.54) and Sensitivity (0.69 and 0.53) on the RENJI and ADAM datasets, respectively.

Conclusions: The proposed method offers a promising solution for accurately segmenting intracranial aneurysms, showcasing superior performance compared to existing approaches. By integrating hybrid loss functions and vessel information, we enhance the model's ability to delineate intricate aneurysm structures, contributing to improved diagnosis and treatment planning for patients with intracranial aneurysms.

Keywords

Aneurysm segmentation / nnU-Net / TOF-MRA

References

[1] Bor AS,Adami J.Risk of subarachnoid haemorrhage according to number of affected relatives: a population based case-control study.Brain2008;131:2662-5

[2] Nieuwkamp DJ,Algra A,de Rooij NK.Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: a meta-analysis.Lancet Neurol2009;8:635-42

[3] Vlak MH,Brandenburg R.Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis.Lancet Neurol2011;10:626-36

[4] Suarez JI,Selman WR.Aneurysmal subarachnoid hemorrhage.N Engl J Med2006;354:387-96

[5] Etminan N.Unruptured intracranial aneurysms: development, rupture and preventive management.Nat Rev Neurol2016;12:699-713

[6] Işın A,Şah M.Review of MRI-based brain tumor image segmentation using deep learning methods.Proc Comput Sci2016;102:317-24

[7] Yu B,Wang L,Zhou L.Medical image synthesis via deep learning. In: Lee G, Fujita H, editors. Deep learning in medical image analysis. Cham: Springer International Publishing; 2020. pp. 23-44.

[8] Haskins G,Yan P.Deep learning in medical image registration: a survey.Mach Vis Appl2020;31:1060

[9] Qu J,Li Y.A deep learning framework for intracranial aneurysms automatic segmentation and detection on magnetic resonance T1 images.Eur Radiol2024;34:2838-48

[10] Zhu G,Yang T.Deep learning-based recognition and segmentation of intracranial aneurysms under small sample size.Front Physiol2022;13:1084202 PMCID:PMC9806214

[11] Yuan W,Guo Y,Xue Q.DCAU-Net: dense convolutional attention U-Net for segmentation of intracranial aneurysm images.Vis Comput Ind Biomed Art2022;5:9 PMCID:PMC8960533

[12] Patel TR,Veeturi SS.Evaluating a 3D deep learning pipeline for cerebral vessel and intracranial aneurysm segmentation from computed tomography angiography-digital subtraction angiography image pairs.Neurosurg Focus2023;54:E13

[13] Shao D,Liu X.3D intracranial aneurysm classification and segmentation via unsupervised dual-branch learning.IEEE J Biomed Health Inform2023;27:1770-9

[14] Sichtermann T,Sijben R,Freiherr J.Deep learning-based detection of intracranial aneurysms in 3D TOF-MRA.AJNR Am J Neuroradiol2019;40:25-32 PMCID:PMC7048599

[15] Di Noto T,Tourbier S.Towards automated brain aneurysm detection in TOF-MRA: open data, weak labels, and anatomical knowledge.Neuroinformatics2023;21:21-34 PMCID:PMC9931814

[16] Ham S,Yun J.Automated detection of intracranial aneurysms using skeleton-based 3D patches, semantic segmentation, and auxiliary classification for overcoming data imbalance in brain TOF-MRA.Sci Rep2023;13:12018 PMCID:PMC10368697

[17] Chen M,Wang D.Deep learning-based segmentation of cerebral aneurysms in 3D TOF-MRA using coarse-to-fine framework.arXiv2021;2110.13432

[18] Timmins KM,Bennink E.Comparing methods of detecting and segmenting unruptured intracranial aneurysms on TOF-MRAS: the ADAM challenge.Neuroimage2021;238:118216

[19] Isensee F,Kohl SAA,Maier-Hein KH.nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation.Nat Methods2021;18:203-11

[20] Wu Z,van den Hengel A.Bridging category-level and instance-level semantic image segmentation.arXiv2016;1605.06885

[21] Lin T,Girshick R.Focal loss for dense object detection. In 2017 IEEE International Conference on Computer Vision (ICCV). 2017; pp. 2999-3007.

[22] Salehi SSM,Gholipour A.Tversky loss function for image segmentation using 3D fully convolutional deep networks. In: Wang Q, Shi Y, Suk H, Suzuki K, editors. Machine learning in medical imaging. Cham: Springer International Publishing; 2017. pp. 379-87.