Implementation of chaotic atom search optimized SWIN transformer architecture for efficient corpus callosum segmentation in brain MRI images

• Autorzy: Avasarala, Padmanabha Sarma , Govindakumar, Saranya

Warianty tytułu

PL

Implementacja zoptymalizowanej architektury transformatora SWIN w zakresie wyszukiwania atomów chaotycznych w celu zapewnienia efektywnej segmentacji ciała modzelowatego w obrazach MRI mózgu

• Języki publikacji: EN

Abstrakty

EN

A crucial component of the brain that facilitates neuronal communication between the two halves of the brain is the corpus callosum (CC). Processing sensorial, motor, and sophisticated intellectual impulses is the major job of the corpus callosum, which integrates and transfers data from both cerebral hemispheres. Segmentation the CC from brain MRIs is a highly challenging technique because of the low brightness of the surrounding organs and tissues. CNN has historically performed better in segmenting medical images, but in 2021 Microsoft researchers created a novel transformer-based structure that outperformed the prior classification methods. As a result, we propose a CC segmentation method based on the Chaotic Atom Search Optimized Swin (Shifted Window) Transformer architecture. The brain MR imaging database is collected using the opensource OASIS platform. Wavelet Thresholding preprocessing compresses the brain MR images and lowers unneeded noise. The corpus callosum is segmented from images of the skull using the proposed Swin framework, which has been developed and trained. The suggested framework is implemented in the Python environment, and metrics like as accuracy, recall, precision, and F1-score are analyzed and compared with existing systems.

PL

Kluczowym elementem mózgu, który ułatwia komunikację neuronalną między dwiema połówkami mózgu, jest ciało modzelowate (CC). Przetwarzanie impulsów czuciowych, motorycznych i wyrafinowanych intelektualnych to główne zadanie ciała modzelowatego, które integruje i przesyła dane z obu półkul mózgowych. Segmentacja CC na podstawie rezonansu magnetycznego mózgu jest techniką bardzo wymagającą ze względu na niską jasność otaczających narządów i tkanek. W przeszłości CNN radziło sobie lepiej w segmentacji obrazów medycznych, ale w 2021 r. badacze firmy Microsoft stworzyli nowatorską strukturę opartą na transformatorach, która przewyższała wcześniejsze metody klasyfikacji. W rezultacie proponujemy metodę segmentacji CC opartą na architekturze transformatora Chaotic Atom Search Optimized Swin (Shifted Window). Baza danych obrazowania MR mózgu jest gromadzona przy użyciu platformy OASIS typu open source. Wstępne przetwarzanie Wavelet Thresholding kompresuje obrazy MR mózgu i obniża niepotrzebne szumy. Ciało modzelowate jest segmentowane na podstawie obrazów czaszki przy użyciu proponowanego modelu Swin, który został opracowany i przeszkolony. Sugerowany framework jest zaimplementowany w środowisku Python, a wskaźniki takie jak dokładność, przypominanie, precyzja i wynik F1 są analizowane i porównywane z istniejącymi systemami.

Słowa kluczowe

PL

ciało modzelowate; próg falkowy; MRI; zoptymalizowane wyszukiwanie atomów chaotycznych; CASO

EN

corpus callosum; wavelet thresholding; MRI; chaotic atom search optimized; CASO; shifted window; SWIN

• Czasopismo: Przegląd Elektrotechniczny
• Rocznik: 2023
• Tom: R. 99, nr 12
• Strony: 185--189
• Opis fizyczny: Bibliogr. 19 poz., rys.

Twórcy

autor

Avasarala, Padmanabha Sarma

• Department of Biomedical Engineering, Vel Tech Rangarajan Dr. SagunthalaR&DInstitute of Science and Technology

autor

Govindakumar, Saranya

• Department of Biomedical Engineering, Vel Tech Rangarajan Dr. SagunthalaR&DInstitute of Science and Technology

Bibliografia

• [1] Shai Berman, Shir Filo, and Aviv A Mezer. Modeling conduction delays in the corpus callosum using mri-measured g-ratio. Neuroimage, 195:128–139, 2019.
• [2] Ángela Bernabéu-Sanz, Sandra Morales, Valery Naranjo, and Ángel P Sempere. Contribution of gray matter atrophy and white matter damage to cognitive impairment in mildly disabled relapsing-remitting multiple sclerosis patients. Diagnostics, 11(3):578, 2021.
• [3] J Blaauw and LC Meiners. The splenium of the corpus callosum: embryology, anatomy, function and imaging with pathophysiological hypothesis. Neuroradiology, 62:563–585, 2020.
• [4] J Blaauw and LC Meiners. The splenium of the corpus callosum: embryology, anatomy, function and imaging with pathophysiological hypothesis. Neuroradiology, 62:563–585, 2020.
• [5] J Blaauw and LC Meiners. The splenium of the corpus callosum: embryology, anatomy, function and imaging with pathophysiological hypothesis. Neuroradiology, 62:563–585, 2020.
• [6] Thais Caldeira, Paulo Rogério Julio, Simone Appenzeller, and Leticia Rittner. inccsight: A software for exploration and visualization of dt-mri data of the corpus callosum. Computers & Graphics, 99:259–271, 2021.
• [7] Aaron Carass, Jennifer L Cuzzocreo, Shuo Han, Carlos R Hernandez-Castillo, Paul E Rasser, Melanie Ganz, Vincent Beliveau, Jose Dolz, Ismail Ben Ayed, Christian Desrosiers, et al. Comparing fully automated state-of-the-art cerebellum parcellation from magnetic resonance images. Neuroimage, 183:150–172, 2018.
• [8] Anjali Chandra, Shrish Verma, Ajay Singh Raghuvanshi, Narendra D Londhe, Narendra Kuber Bodhey, and Kumar Subham. Corpus callosum segmentation from brain mri and its possible application in detection of diseases. In 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), pages 1–4. IEEE, 2019.
• [9] GS Cover, William Garcia Herrera, Mariana P Bento, Simone Appenzeller, and Letícia Rittner. Computational methods for corpus callosum segmentation on mri: A systematic literature review. Computer methods and programs in biomedicine, 154:25–35, 2018.
• [10] GS Cover, William Garcia Herrera, Mariana P Bento, Simone Appenzeller, and Letícia Rittner. Computational methods for corpus callosum segmentation on mri: A systematic literature review. Computer methods and programs in biomedicine, 154:25–35, 2018.
• [11] Chitradevi Dhakhinamoorthy, Sathish Kumar Mani, Sandeep Kumar Mathivanan, Senthilkumar Mohan, Prabhu Jayagopal, Saurav Mallik, and Hong Qin. Hybrid whale and gray wolf deep learning optimization algorithm for prediction of alzheimer’s disease. Mathematics, 11(5):1136, 2023.
• [12] Douglas N Greve, Benjamin Billot, Devani Cordero, Andrew Hoopes, Malte Hoffmann, Adrian V Dalca, Bruce Fischl, Juan Eugenio Iglesias, and Jean C Augustinack. A deep learning toolbox for automatic segmentation of subcortical limbic structures from mri images. Neuroimage, 244:118610, 2021.
• [13] Fidel Hernandez, Chiara Giordano, Maged Goubran, Sherveen Parivash, Gerald Grant, Michael Zeineh, and David Camarillo. Lateral impacts correlate with falx cerebri displacement and corpus callosum trauma in sports-related concussions. Biomechanics and modeling in mechanobiology, 18:631–649, 2019.
• [14] Amanda K Huber, David A Giles, Benjamin M Segal, and David N Irani. An emerging role for eotaxins in neurodegenerative disease. Clinical Immunology, 189:29–33, 2018.
• [15] Paulo Rogério Julio, Thais Caldeira, Gustavo Retuci Pinheiro, Carla Helena Capello, Renan Bazuco Fritolli, Roberto Marini, Fernando Cendes, Paula Teixeira Fernandes, Lilian TL Costallat, Leticia Rittner, et al. Microstructural changes in the corpus callosum in systemic lupus erythematous. Cells, 12(3):355, 2023.
• [16] Mónica López-Vicente, Sander Lamballais, Suzanne Louwen, Manon Hillegers, Henning Tiemeier, Ryan L Muetzel, and Tonya White. White matter microstructure correlates of age, sex, handedness and motor ability in a population-based sample of 3031 school-age children. Neuroimage, 227:117643, 2021.
• [17] Siuly Siuly and Yanchun Zhang. Medical big data: neurological diseases diagnosis through medical data analysis. Data Science and Engineering, 1:54–64, 2016.
• [18] Stefanus E Sugijono, Rahmad Mulyadi, Salsabila Firdausia, Joedo Prihartono, and Riwanti Estiasari. Corpus callosum index correlates with brain volumetry and disability in multiple sclerosis patients. Neurosciences Journal, 25(3):193–199, 2020.
• [19] Lara M Wierenga, Marieke GN Bos, Elisabeth Schreuders, Ferdi vd Kamp, Jiska S Peper, Christian K Tamnes, and Eveline A Crone. Unraveling age, puberty and testosterone effects on subcortical brain development across adolescence. Psychoneuroendocrinology, 91:105–114, 2018.

Identyfikatory

DOI

10.15199/48.2023.12.33