Araştırma Makalesi
BibTex RIS Kaynak Göster

Intracranial Hemorrhage Identification from Computed Tomography Images Using a Lightweight Deep Learning Model

Yıl 2024, Cilt: 14 Sayı: 2, 384 - 395, 31.12.2024
https://doi.org/10.54370/ordubtd.1545060

Öz

Intracranial hemorrhage, which is defined as blood leakage into the brain tissue, is a neurological complication that requires urgent medical intervention. Accordingly, early diagnosis of the hemorrhage significantly affects the chance of survival of patients and the recovery process. The presence of intracranial hemorrhage and its location can be identified quickly and effectively by deep learning-based approaches from computed tomography (CT) or magnetic resonance (MR) images, which radiologists commonly prefer to diagnose intracranial hematomas. These methods can significantly reduce the workload of radiologists and help make more accurate detections in complex cases. Accordingly, potential deaths or physical dysfunctions due to the hemorrhage can be prevented. This study proposes a CNN-based lightweight deep learning model that can accurately detect and classify intracranial hemorrhage from computed tomography images. Comparative experimental analyses with popular CNN models such as DenseNet121, MobileNet, and Inception V1 have shown that the proposed model significantly reduces training time and performs better.

Kaynakça

  • Ahmed, S. N. ve Prakasam, P. (2023). A systematic review on intracranial aneurysm and hemorrhage detection using machine learning and deep learning techniques. Progress in Biophysics and Molecular Biology, 183, 1-16. https://doi.org/10.1016/j.pbiomolbio.2023.07.001
  • Altıntaş, M. (2021). Bilgisayarlı tomografi görüntülerinde inmenin farklı derin öğrenme modelleri ile sınıflandırılması [Yayımlanmamış yüksek lisans tezi]. Necmettin Erbakan Üniversitesi.
  • Burduja M., Ionescu R. T. ve Verga N. (2020). Accurate and efficient intracranial hemorrhage detection and subtype classification in 3D CT scans with convolutional and long short-term memory neural networks, Sensors, 20(19), 1-21. https://doi.org/10.3390/s20195611
  • Champawat, Y. S., Shagun ve Prakash, C. (2023). Literature review for automatic detection and classification of intracranial brain hemorrhage using computed tomography scans. In H. Muthusamy, J. Botzheim ve R. Nayak (Ed.), Lecture notes in electrical engineering: Vol. 1009. Robotics, control and computer vision (s. 39-65). Springer. https://doi.org/10.1007/978-981-99-0236-1_4
  • Cordonnier C., Demchuk A., Ziai W. ve Anderson C. S. (2018). Intracerebral haemorrhage: current approaches to acute management. The Lancet, 392(10154), 1257-1268. https://doi.org/10.1016/S0140-6736(18)31878-6
  • Desai V., Flanders A. E. ve Lakhani P. (2017). Application of deep learning in neuroradiology: automated detection of basal ganglia hemorrhage using 2D-convolutional neural networks. arXiv preprint arXiv, 1-7. https://doi.org/10.48550/arXiv.1710.03823
  • Gautam A. ve Raman B. (2021). Towards effective classification of brain hemorrhagic and ischemic stroke using CNN. Biomedical Signal Processing Control, 63, 1-13. https://doi.org/10.1016/j.bspc.2020.102178
  • Gençtürk T. H., Gülağız F. K. ve Kaya İ. (2023). Derin öğrenme yöntemleri kullanılarak BT taramalarında beyin kanaması teşhisinin karşılaştırmalı bir analizi. Zeki Sistemler Teori ve Uygulamaları Dergisi, 6( 1), 75–84. https://doi.org/10.38016/jista.1215025
  • Grandini, M., Bagli, E. ve Visani, G. (2020). Metrics for multi-class classification: An overview. arXiv preprint arXiv, 1-17. https://doi.org/10.48550/arXiv.2008.05756
  • Howard A. G., Zhu M., Chen B., Kalenichenko D., Wang W., Weyand T., Andreetto M. ve Adam H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv, 1-9. https://doi.org/10.48550/arXiv.1704.04861
  • Huang G., Liu Z., Maaten L. V. D. ve Weinberger K. Q. (2017, Temmuz, 21-26). Densely connected convolutional networks [Sözlü sunum]. IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA. https://doi.org/10.1109/CVPR.2017.243
  • Jelassi, S. ve Li, Y. (2022, Temmuz, 17-23). Towards understanding how momentum improves generalization in deep learning [Sözlü sunum]. International Conference on Machine Learning, Maryland, USA. https://proceedings.mlr.press/v162/jelassi22a/jelassi22a.pdf
  • Ker J., Singh S. P., Bai Y., Rao J., Lim T. ve Wang L. (2019). Image thresholding improves 3-dimensional convolutional neural network diagnosis of different acute brain hemorrhages on computed tomography scans. Sensors, 19( 9), 1-12. https://doi.org/10.3390/s19092167
  • Mushtaq M. F., Shahroz M., Aseere A. M., Shah H., Majeed R., Shehzad D. ve Samad A. (2021). BHCNet: neural network-based brain hemorrhage classification using head CT scan. IEEE Access, 9, 113901-113916. https://doi.org/10.1109/ACCESS.2021.3102740
  • Nilsson O. G., Lindgren A., Stahl N., Brandt L. ve Saveland H. (2000). Incidence of intracerebral and subarachnoid haemorrhage in southern Sweden. Journal of Neurology, Neurosurg & Psychiatry, 69(5), 601-607. https://doi.org/10.1136/jnnp.69.5.601
  • Phong T. D., Duong H., Nguyen H. T., Trong N. T., Nguyen V. H., Hoa T. V. ve Snasel V. (2017, Ocak, 13-16). Brain hemorrhage diagnosis by using deep learning. International Conference on Machine Learning and Soft Computing, New York, NY, USA. https://doi.org/10.1145/3036290.3036326
  • Polat Ö. ve Kartal M. S. (2023). Derin öğrenme ile pencere ayarlı görüntüler kullanılarak beyin inme segmentasyon performansının geliştirilmesi. Gümüşhane Üniversitesi Fen Bilimleri Dergisi, 13(4), 1094-1109. https://doi.org/10.17714/gumusfenbil.1319024
  • Radiological Society of North America (2019). RSNA intracranial hemorrhage detection challenge. https://www.rsna.org/rsnai/ai-image-challenge/rsna-intracranial-hemorrhage-detection-challen ge-2019 adresinden 2 Ağustos 2024 tarihinde alınmıştır.
  • Rane H. ve Warhade K. (2021, Mart, 5-7). A survey on deep learning for intracranial hemorrhage detection [Sözlü sunum]. International Conference on Emerging Smart Computing and Informatics, Pune, India. https://doi.org/10.1109/ESCI50559.2021.9397009
  • Sage A. ve Badura P. (2020). Intracranial hemorrhage detection in head CT using double-branch convolutional neural network, support vector machine, and random forest. Applied Sciences, 10(21), 1-12. https://doi.org/10.3390/app10217577
  • Salehinejad H., Kitamura J., Ditkofsky N., Lin A., Bharatha A., Suthiphosuwan S., Lin H., Wilson J. R., Mamdani M. ve Colak E. (2021). A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography. Scientific Reports, 11(1), 1-11. https://doi.org/10.1038/s41598-021-95533-2
  • Szegedy C., Vanhoucke V., Ioffe S., Shlens J. ve Wojna Z. (2016, Haziran, 27-30). Rethinking the inception architecture for computer vision [Poster sunumu]. IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA. https://doi.org/10.1109/CVPR.2016.308
  • Wang X., Shen T., Yang S. ve Lan J. (2021). A deep learning algorithm for automatic detection and classification of acute intracranial hemorrhages in head CT scans. Neuroimage Clinical, 32, 1-10. https://doi.org/10.1016/j.nicl.2021.102785
  • Yalçın S. ve Vural H. (2022). Brain stroke classification and segmentation using encoder-decoder based deep convolutional neural networks. Computers in Biology and Medicine, 149, 1-14. https://doi.org/10.1016/j.compbiomed.2022.105941
  • Yeo M., Tahayori B., Kok H. K., Maingard J., Kutaiba N., Russell J., Thijs V., Jhamb , R. V. Chandra A., Brooks M., Barras C. D. ve Asadi H. (2021). Review of deep learning algorithms for the automatic detection of intracranial hemorrhages on computed tomography head imaging. Journal of NeuroInterventional Surgery, 13, 369-378. https://doi.org/10.1136/neurintsurg-2020-017099
  • Zhang M., Gu S. ve Shi Y. (2022). The use of deep learning methods in low-dose computed tomography image reconstruction: A systematic review. Complex and Intelligent Systems, 8, 5545-5561. https://doi.org/10.1007/s40747-022-00724-7

Hafif Bir Derin Öğrenme Modeli İle Bilgisayarlı Tomografi Görüntülerinden Beyin Kanaması Tespiti

Yıl 2024, Cilt: 14 Sayı: 2, 384 - 395, 31.12.2024
https://doi.org/10.54370/ordubtd.1545060

Öz

Beyin dokusu içine kan sızması durumu olarak ifade edilen beyin kanaması, acil tıbbi müdahale gerektiren nörolojik bir komplikasyondur. Bu sebeple, beyin kanamasında erken tanı, hastaların hayatta kalma şansını ve iyileşme sürecini önemli ölçüde etkiler. Beyin kanaması teşhisinde, radyologlarca yaygın olarak tercih edilen bilgisayarlı tomografi (BT) ve manyetik rezonans (MR) görüntüleri, derin öğrenme tabanlı yaklaşımlar ile analiz edilerek, beyin kanamasının varlığı ve kanamanın yeri hızlı ve etkili bir şekilde tespit edilebilir. Bu yöntemler, radyologların iş yükünü önemli ölçüde azaltabileceği gibi, kompleks vakalarda daha kesin teşhisler koyulmasına da yardımcı olabilir. Buna bağlı olarak, beyin kanaması kaynaklı ölümlerin veya bedensel işlev bozukluklarının önüne geçilebilir. Bu çalışmada, bilgisayarlı tomografi görüntüleri üzerinden beyin kanaması ve türünü yüksek doğrulukta tespit edebilen CNN tabanlı düşük boyutlu bir derin öğrenme modeli önerilmiştir. DenseNet121, MobileNet ve Inception V1 gibi popüler CNN modelleri ile yapılan karşılaştırmalı deneysel analizler, önerilen modelin, eğitim süresini önemli ölçüde kısalttığını ve daha başarılı bir performans sergilediğini göstermiştir.

Etik Beyan

Bu makalenin yayınlanmasıyla ilgili herhangi bir etik sorun bulunmamaktadır.

Teşekkür

Bu çalışma 2209-A - Üniversite Öğrencileri Araştırma Projeleri Destekleme Programı kapsamında “Bilgisayarlı Tomografi Görüntüleri Kullanılarak Derin Öğrenme ile Beyin Kanaması Tespiti” isimli proje ile TÜBİTAK tarafından desteklenmiştir.

Kaynakça

  • Ahmed, S. N. ve Prakasam, P. (2023). A systematic review on intracranial aneurysm and hemorrhage detection using machine learning and deep learning techniques. Progress in Biophysics and Molecular Biology, 183, 1-16. https://doi.org/10.1016/j.pbiomolbio.2023.07.001
  • Altıntaş, M. (2021). Bilgisayarlı tomografi görüntülerinde inmenin farklı derin öğrenme modelleri ile sınıflandırılması [Yayımlanmamış yüksek lisans tezi]. Necmettin Erbakan Üniversitesi.
  • Burduja M., Ionescu R. T. ve Verga N. (2020). Accurate and efficient intracranial hemorrhage detection and subtype classification in 3D CT scans with convolutional and long short-term memory neural networks, Sensors, 20(19), 1-21. https://doi.org/10.3390/s20195611
  • Champawat, Y. S., Shagun ve Prakash, C. (2023). Literature review for automatic detection and classification of intracranial brain hemorrhage using computed tomography scans. In H. Muthusamy, J. Botzheim ve R. Nayak (Ed.), Lecture notes in electrical engineering: Vol. 1009. Robotics, control and computer vision (s. 39-65). Springer. https://doi.org/10.1007/978-981-99-0236-1_4
  • Cordonnier C., Demchuk A., Ziai W. ve Anderson C. S. (2018). Intracerebral haemorrhage: current approaches to acute management. The Lancet, 392(10154), 1257-1268. https://doi.org/10.1016/S0140-6736(18)31878-6
  • Desai V., Flanders A. E. ve Lakhani P. (2017). Application of deep learning in neuroradiology: automated detection of basal ganglia hemorrhage using 2D-convolutional neural networks. arXiv preprint arXiv, 1-7. https://doi.org/10.48550/arXiv.1710.03823
  • Gautam A. ve Raman B. (2021). Towards effective classification of brain hemorrhagic and ischemic stroke using CNN. Biomedical Signal Processing Control, 63, 1-13. https://doi.org/10.1016/j.bspc.2020.102178
  • Gençtürk T. H., Gülağız F. K. ve Kaya İ. (2023). Derin öğrenme yöntemleri kullanılarak BT taramalarında beyin kanaması teşhisinin karşılaştırmalı bir analizi. Zeki Sistemler Teori ve Uygulamaları Dergisi, 6( 1), 75–84. https://doi.org/10.38016/jista.1215025
  • Grandini, M., Bagli, E. ve Visani, G. (2020). Metrics for multi-class classification: An overview. arXiv preprint arXiv, 1-17. https://doi.org/10.48550/arXiv.2008.05756
  • Howard A. G., Zhu M., Chen B., Kalenichenko D., Wang W., Weyand T., Andreetto M. ve Adam H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv, 1-9. https://doi.org/10.48550/arXiv.1704.04861
  • Huang G., Liu Z., Maaten L. V. D. ve Weinberger K. Q. (2017, Temmuz, 21-26). Densely connected convolutional networks [Sözlü sunum]. IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA. https://doi.org/10.1109/CVPR.2017.243
  • Jelassi, S. ve Li, Y. (2022, Temmuz, 17-23). Towards understanding how momentum improves generalization in deep learning [Sözlü sunum]. International Conference on Machine Learning, Maryland, USA. https://proceedings.mlr.press/v162/jelassi22a/jelassi22a.pdf
  • Ker J., Singh S. P., Bai Y., Rao J., Lim T. ve Wang L. (2019). Image thresholding improves 3-dimensional convolutional neural network diagnosis of different acute brain hemorrhages on computed tomography scans. Sensors, 19( 9), 1-12. https://doi.org/10.3390/s19092167
  • Mushtaq M. F., Shahroz M., Aseere A. M., Shah H., Majeed R., Shehzad D. ve Samad A. (2021). BHCNet: neural network-based brain hemorrhage classification using head CT scan. IEEE Access, 9, 113901-113916. https://doi.org/10.1109/ACCESS.2021.3102740
  • Nilsson O. G., Lindgren A., Stahl N., Brandt L. ve Saveland H. (2000). Incidence of intracerebral and subarachnoid haemorrhage in southern Sweden. Journal of Neurology, Neurosurg & Psychiatry, 69(5), 601-607. https://doi.org/10.1136/jnnp.69.5.601
  • Phong T. D., Duong H., Nguyen H. T., Trong N. T., Nguyen V. H., Hoa T. V. ve Snasel V. (2017, Ocak, 13-16). Brain hemorrhage diagnosis by using deep learning. International Conference on Machine Learning and Soft Computing, New York, NY, USA. https://doi.org/10.1145/3036290.3036326
  • Polat Ö. ve Kartal M. S. (2023). Derin öğrenme ile pencere ayarlı görüntüler kullanılarak beyin inme segmentasyon performansının geliştirilmesi. Gümüşhane Üniversitesi Fen Bilimleri Dergisi, 13(4), 1094-1109. https://doi.org/10.17714/gumusfenbil.1319024
  • Radiological Society of North America (2019). RSNA intracranial hemorrhage detection challenge. https://www.rsna.org/rsnai/ai-image-challenge/rsna-intracranial-hemorrhage-detection-challen ge-2019 adresinden 2 Ağustos 2024 tarihinde alınmıştır.
  • Rane H. ve Warhade K. (2021, Mart, 5-7). A survey on deep learning for intracranial hemorrhage detection [Sözlü sunum]. International Conference on Emerging Smart Computing and Informatics, Pune, India. https://doi.org/10.1109/ESCI50559.2021.9397009
  • Sage A. ve Badura P. (2020). Intracranial hemorrhage detection in head CT using double-branch convolutional neural network, support vector machine, and random forest. Applied Sciences, 10(21), 1-12. https://doi.org/10.3390/app10217577
  • Salehinejad H., Kitamura J., Ditkofsky N., Lin A., Bharatha A., Suthiphosuwan S., Lin H., Wilson J. R., Mamdani M. ve Colak E. (2021). A real-world demonstration of machine learning generalizability in the detection of intracranial hemorrhage on head computerized tomography. Scientific Reports, 11(1), 1-11. https://doi.org/10.1038/s41598-021-95533-2
  • Szegedy C., Vanhoucke V., Ioffe S., Shlens J. ve Wojna Z. (2016, Haziran, 27-30). Rethinking the inception architecture for computer vision [Poster sunumu]. IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA. https://doi.org/10.1109/CVPR.2016.308
  • Wang X., Shen T., Yang S. ve Lan J. (2021). A deep learning algorithm for automatic detection and classification of acute intracranial hemorrhages in head CT scans. Neuroimage Clinical, 32, 1-10. https://doi.org/10.1016/j.nicl.2021.102785
  • Yalçın S. ve Vural H. (2022). Brain stroke classification and segmentation using encoder-decoder based deep convolutional neural networks. Computers in Biology and Medicine, 149, 1-14. https://doi.org/10.1016/j.compbiomed.2022.105941
  • Yeo M., Tahayori B., Kok H. K., Maingard J., Kutaiba N., Russell J., Thijs V., Jhamb , R. V. Chandra A., Brooks M., Barras C. D. ve Asadi H. (2021). Review of deep learning algorithms for the automatic detection of intracranial hemorrhages on computed tomography head imaging. Journal of NeuroInterventional Surgery, 13, 369-378. https://doi.org/10.1136/neurintsurg-2020-017099
  • Zhang M., Gu S. ve Shi Y. (2022). The use of deep learning methods in low-dose computed tomography image reconstruction: A systematic review. Complex and Intelligent Systems, 8, 5545-5561. https://doi.org/10.1007/s40747-022-00724-7
Toplam 26 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular Görüntü İşleme, Derin Öğrenme, Sağlıkta Bilgi İşleme
Bölüm Araştırma Makaleleri
Yazarlar

Emine Betül Altun 0009-0006-1408-3371

Sümeyye Engin 0009-0009-4534-2139

Esma Başkaya 0009-0000-6317-7952

Fatmanur Şafak 0009-0003-3685-311X

Saffet Vatansever 0000-0002-4680-1263

Yayımlanma Tarihi 31 Aralık 2024
Gönderilme Tarihi 7 Eylül 2024
Kabul Tarihi 26 Kasım 2024
Yayımlandığı Sayı Yıl 2024 Cilt: 14 Sayı: 2

Kaynak Göster

APA Altun, E. B., Engin, S., Başkaya, E., Şafak, F., vd. (2024). Hafif Bir Derin Öğrenme Modeli İle Bilgisayarlı Tomografi Görüntülerinden Beyin Kanaması Tespiti. Ordu Üniversitesi Bilim Ve Teknoloji Dergisi, 14(2), 384-395. https://doi.org/10.54370/ordubtd.1545060