PC-MRG'de Biyolojik Sıvı Akış Dinamiği Kullanılarak Ekstraventriküler Beyin Bölgelerinin Denetimsiz Segmentasyonu

Öz

Beyin Omurilik Sıvısı (BOS), sağlığın sürdürülmesinde kritik bir rol oynamaktadır; başta kardiyak döngü ve solunum tarafından yönlendirilen bu sıvının pulsatil hareketi değerli bir fizyolojik sinyal niteliği taşımakta olup anormallikleri giderek artan biçimde nörolojik bozukluklarla ilişkilendirilmektedir. Faz-kontrast manyetik rezonans görüntüleme (PC-MRG), hem anatomik yapıyı hem de akış özelliklerini non-invaziv biçimde yakalamaya olanak tanımaktadır. Bu çalışmada, ön komissür–arka komissür (AK–PK) hattı boyunca elde edilen PC-MRG verilerinden türetilen BOS akış dinamiklerini kullanarak ekstraventrikler beyin bölgelerini bölütlemeye yönelik denetimsiz bir yöntem önerilmektedir. PC-MRG verileri, kardiyak döngü başına 23–30 görüntü olacak şekilde sekiz sağlıklı gönüllüden (VENC = 10 cm/s) toplanmıştır. Zaman, frekans ve dalgacık (wavelet) domenlerinden elde edilen gösterimlerden çıkarılan voksel bazlı özellikler K-ortalamalar algoritması ile kümelenmiştir. Ablasyon çalışması, frekans domenine ait özelliklerin en ayrıştırıcı domain olduğunu doğrulamıştır. Uzman nöroradyologlar tarafından doğrulanan segmentler, ventrikül ötesindeki anatomik bölgelerle — gri madde, beyaz madde, vasküler yapılar ve BOS boşlukları — güçlü bir uyum sergilemiştir. Ayrıca segmente edilen bölgelerdeki BOS akış özellikleri de karşılaştırmalı analiz amacıyla incelenmiş ve sunulmuştur. Ekstraventrikler bölgelere odaklanılarak tüm beyin genelinde BOS dinamiklerinin daha kapsamlı biçimde incelenmesi mümkün kılınmış; BOS’un çevre dokularla nasıl etkileşime girdiğine dair yeni bulgular ortaya konmuştur. Beyin genelinde uygulanan bu etiket-bağımsız yaklaşım, BOS akışının bozulduğu nörolojik durumların daha erken tespitine ve daha etkin izlenmesine zemin hazırlamakta; glimatik taşıma, nöroinflamasyon ve sıvı-doku etkileşimlerinin rolüne yönelik gelecekteki araştırmalar için temel oluşturmaktadır.

Anahtar Kelimeler

• PC-MRI
• Beyin Omurilik Sıvısı
• Gözetimsiz Segmentasyon
• K-means Kümeleme
• Ekstraventriküler Nörogörüntüleme
• Dalgacık Analizi

Etik Beyan

Bilkent Üniversitesi İnsan Araştırmaları Etik Kurulu tarafından etik onay verilmiştir (2024-610) ve görüntüleme işleminden önce tüm katılımcılardan yazılı bilgilendirilmiş onam alınmıştır.

Unsupervised Segmentation of Extraventricular Brain Regions in PC-MRI Using CSF Flow Dynamics

Öz

Cerebrospinal fluid (CSF) plays a crucial role in maintaining health, and its pulsatile movement driven primarily by the cardiac cycle and respiration serves as a valuable physiological signal, with abnormalities increasingly associated with neurological disorders. Phase-contrast magnetic resonance imaging (PC-MRI) offers a non-invasive means of capturing both anatomical structure and flow characteristics. In this study, we propose an unsupervised method to segment extraventricular brain regions using CSF flow dynamics derived from PC-MRI data acquired along the anterior commissure–posterior commissure (AC–PC) line. PC-MRI data were collected from eight healthy volunteers (VENC = 10 cm/s) with 23–30 images per cardiac cycle. Voxel-wise features — extracted from time-, frequency-, and wavelet-domain representations —were clustered using the K-means algorithm. An ablation study confirmed that frequency-domain features are the most discriminative single domain. The resulting segments, validated by expert neuroradiologists, showed strong alignment with anatomical regions- grey matter, white matter, vasculature, and CSF spaces- beyond the ventricles. Additionally, the CSF flow characteristics within the segmented regions were also examined and presented for comparative analysis. By focusing on extraventricular regions, this method enables a more comprehensive analysis of CSF dynamics across the entire brain, offering new insights into how CSF interacts with surrounding tissues. This brain-wide, label-free approach opens the door to earlier detection and improved monitoring of neurological conditions where CSF flow is disrupted and lays the groundwork for future research into the role of glymphatic transport, neuroinflammation, and fluid-tissue interactions in both health and disease.

Anahtar Kelimeler

• PC-MRI
• Cerebrospinal fluid
• Unsupervised segmentation
• K-means clustering
• Extraventricular neuroimaging
• Wavelet analysis

Etik Beyan

Ethical approval was granted by Bilkent University Human Research Ethics Committee (2024-610) and written informed consent was obtained from all participants prior to imaging.

Kaynakça

1. Abdullah, A., Hirayama, A., Yatsushiro, S., Matsumae, M., & Kuroda, K. (2013). Cerebrospinal fluid image segmentation using spatial fuzzy clustering method with improved evolutionary expectation maximization. 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC),
2. Alperin, N., & Lee, S. H. (2003). PUBS: pulsatility‐based segmentation of lumens conducting non‐steady flow. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine, 49(5), 934-944.
3. Aslan, E., & Özüpak, Y. (2025). Performance comparison of deep learning models in brain tumor classification. Balkan Journal of Electrical and Computer Engineering, 13(2), 203-209. Battal, B., Kocaoglu, M., Bulakbasi, N., Husmen, G., Tuba Sanal, H., & Tayfun, C. (2011). Cerebrospinal fluid flow imaging by using phase-contrast MR technique. The British journal of radiology, 84(1004), 758-765.
4. Bayram, B., Kunduracioglu, I., İnce, S., & Pacal, I. (2025). A systematic review of deep learning in MRI-based cerebral vascular occlusion-based brain diseases. Neuroscience. https://doi.org/10.1016/j.neuroscience.2025.01.020
5. Choi, S.-H., Chi, J.-G., Kim, Y.-B., & Cho, Z.-H. (2013). Anterior commissure-posterior commissure revisited. Korean journal of radiology, 14(4), 653-661.
6. Ikotun, A. M., Ezugwu, A. E., Abualigah, L., Abuhaija, B., & Heming, J. (2023). K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data. Information Sciences, 622, 178-210.
7. Karki, P., Murphy, M. C., Ganji, S., Gunter, J. L., Graff‐Radford, J., Jones, D. T., Botha, H., Cutsforth‐Gregory, J. K., Elder, B. D., & Jack Jr, C. R. (2025). Real‐Time 2D Phase‐Contrast MRI to Assess Cardiac‐and Respiratory‐Driven CSF Movement in Normal Pressure Hydrocephalus. Journal of Neuroimaging, 35(1), e70000.
8. Kartal, M. G., & Algin, O. (2014). Evaluation of hydrocephalus and other cerebrospinal fluid disorders with MRI: An update. Insights into imaging, 5, 531-541.

9. Keles, A., Algin, O., Ozisik, P. A., ŞEN, B., & Vehbi Çelebi, F. (2023). Segmentation of spinal subarachnoid lumen with 3D attention U-Net. Journal of Mechanics in Medicine and Biology, 23(04), 2340011.
10. Kikuta, J., Nagai, Y., Kamagata, K., Chen, H., Yamada, Y., Kitagawa, T., Uchida, W., Iseki, R., Guo, S., & Takabayashi, K. (2025). Association between frontal pole atrophy and glymphatic dysfunction in patients with bipolar disorder. Journal of Affective Disorders, 119686.
11. Kim, D.-S., Choi, J.-U., Huh, R., Yun, P.-H., & Kim, D.-I. (1999). Quantitative assessment of cerebrospinal fluid hydrodynamics using a phase-contrast cine MR image in hydrocephalus. Child's Nervous System, 15, 461-467.
12. Korbecki, A., Zimny, A., Podgórski, P., Sąsiadek, M., & Bladowska, J. (2019). Imaging of cerebrospinal fluid flow: fundamentals, techniques, and clinical applications of phase-contrast magnetic resonance imaging. Polish journal of radiology, 84, 240-250.
13. Lokossou, A., Balédent, O., Garnotel, S., Page, G., Balardy, L., Czosnyka, Z., Payoux, P., & Schmidt, E. (2018). ICP monitoring and phase-contrast MRI to investigate intracranial compliance. Intracranial Pressure & Neuromonitoring XVI,
14. Menick, B. J. (2001). Phase-contrast magnetic resonance imaging of cerebrospinal fluid flow in the evaluation of patients with Chiari I malformation. Neurosurgical Focus, 11(1), 1-4.
15. Oti, E. U., Olusola, M. O., Eze, F. C., & Enogwe, S. U. (2021). Comprehensive review of K-Means clustering algorithms. criterion, 12, 22-23.
16. Patel, R. P., Sitton, C. W., Ketonen, L. M., Hou, P., Johnson, J. M., Romo, S., Fletcher, S., Shah, M. N., Kerr, M., & Zaky, W. (2018). Phase-contrast cerebrospinal fluid flow magnetic resonance imaging in qualitative evaluation of patency of CSF flow pathways prior to infusion of chemotherapeutic and other agents into the fourth ventricle. Child's Nervous System, 34, 481-486.
17. Stadlbauer, A., Salomonowitz, E., Van Der Riet, W., Buchfelder, M., & Ganslandt, O. (2010). Insight into the patterns of cerebrospinal fluid flow in the human ventricular system using MR velocity mapping. Neuroimage, 51(1), 42-52.
18. Sundareswaran, K. S., Frakes, D. H., Fogel, M. A., Soerensen, D. D., Oshinski, J. N., & Yoganathan, A. P. (2009). Optimum fuzzy filters for phase‐contrast magnetic resonance imaging segmentation. Journal of Magnetic Resonance Imaging: An Official Journal of the International Society for Magnetic Resonance in Medicine, 29(1), 155-165.
19. Tsou, C.-H., Cheng, Y.-C., Huang, C.-Y., Chen, J.-H., Chen, W.-H., Chai, J.-W., & Chen, C. C.-C. (2021). Using deep learning convolutional neural networks to automatically perform cerebral aqueduct CSF flow analysis. Journal of Clinical Neuroscience, 90, 60-67.
20. Weickert, J. (1998). Anisotropic diffusion in image processing (Vol. 1). Teubner Stuttgart.
21. Weltens, C., Menten, J., Feron, M., Bellon, E., Demaerel, P., Maes, F., Van den Bogaert, W., & van der Schueren, E. (2001). Interobserver variations in gross tumor volume delineation of brain tumors on computed tomography and impact of magnetic resonance imaging. Radiotherapy and Oncology, 60(1), 49-59.
22. Wu, L., Wang, S., Liu, J., Hou, L., Li, N., Su, F., Yang, X., Lu, W., Qiu, J., & Zhang, M. (2025). A survey of MRI-based brain tissue segmentation using deep learning. Complex & Intelligent Systems, 11(1), 1-16.
23. Yoshida, K., Takahashi, H., Saijo, M., Ueguchi, T., Tanaka, H., Fujita, N., & Murase, K. (2009). Phase-contrast MR studies of CSF flow rate in the cerebral aqueduct and cervical subarachnoid space with correlation-based segmentation. Magnetic Resonance in Medical Sciences, 8(3), 91-100.