AI for Glioma with Multimodal Data - Histopathology, Clinical, Genomics,...
Abstract:
Computational pathology using computer vision techniques are required to standardize the disease stage evaluation for Glioblastoma multiforme (GBM). In the following works, we perform:
Disease stage classification [1-5, 7]
The performance of various feature descriptors (Nuclei/Texture/Color/Deep-features) for disease stage of gliomas [1-4]
Subtype clustering/classification of low grade and high grade gliomas [6, 9, 10, 11]
AI models for predicting gene mutations (e.g. IDH1) from H & E + clinical data [8]
We test the performance of various feature descriptors - hand-tuned, ML, DL features. We develop, and benchmark various deep learning models, frameworks - MIL, ABMIL, and multimodal fusion approaches.
Disease Stage Classification and Significance of Features [1]
Subtype Classification with Ensemble Fusion AI (EFAI) [9]
References:
[1] K. Fukuma, H. Kawanaka, V. B. S. Prasath, B. J. Aronow, H. Takase. Feature extraction and disease stage classification for glioma histopathology images. IEEE 17th International Conference on e-Health Networking, Applications and Services (Healthcom), Boston, USA. Proc. IEEE, pp. 598-599, Oct 2015. doi:10.1109/HealthCom.2015.7454574
[2] K. Fukuma, V. B. S. Prasath, H. Kawanaka, B. J. Aronow, H. Takase. A study on feature extraction and disease stage classification for glioma pathology images. IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Vancouver, Canada. Proc. IEEE, pp. 2150-2156, July 2016. doi:10.1109/FUZZ-IEEE.2016.7737958
[3] K. Fukuma, V. B. S. Prasath, H. Kawanaka, B. J. Aronow, H. Takase. A study on nuclei segmentation, feature extraction and disease stage classification for human brain histopathological images. International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES), York, UK, Sep 2016. Procedia Computer Science, 96, pp. 1202-1020, Sep 2016. doi:10.1016/j.procs.2016.08.164
[4] A. Yonekura, H. Kawanaka, V. B. S. Prasath, B. J. Aronow, H. Takase. Glioblastoma Multiforme tissue histopathology images based disease stage classification with deep CNN. 6th International Conference on Informatics, Electronics & Vision (ICIEV), Himeji, Hyogo, Japan. Proc. IEEE, September 2017. doi:10.1109/ICIEV.2017.8338558 (Best Paper Award Finalist)
[5] A. Yonekura, H. Kawanaka, V. B. S. Prasath, B. J. Aronow, H. Takase. Improving the generalization of disease stage classification with deep CNN for glioma histopathological images. IEEE International Conference on Bioinformatics and Biomedicine (BIBM), International Workshop on Deep Learning in Bioinformatics, Biomedicine, and Healthcare Informatics (DLB2H), Kansas, MO, USA. Proc. IEEE, pp. 1222-1226, November 2017. doi:10.1109/BIBM.2017.8217831
[6] A. Yonekura, H. Kawanaka, V. B. S. Prasath, B. J. Arronow, S. Tsuruoka. Glioma subtypes clustering method using histopathological image analysis. 7th International Conference on Informatics, Electronics and Vision (ICIEV), and 2nd International Conference on Imaging, Vision and Pattern Recognition (icIVPR), Fukuoka, Japan. Proc. IEEE, pp. 442-446, June 2018. doi:10.1109/ICIEV.2018.8641031
[7] A. Yonekura, H. Kawanaka, V. B. S. Prasath, B. J. Aronow, H. Takase. Automatic disease stage classification of glioblastoma multiforme histopathological images using deep convolutional neural network. Biomedical Engineering Letters, 8(3), 321-327, August 2018. doi:10.1007/s13534-018-0077-0
[8] R. Nakagaki, S. S. Debsarkar, H. Kawanaka, B. Aronow, V. B. S. Prasath. Deep learning-based IDH1 gene mutation prediction using histopathological imaging and clinical data. Computers in Biology and Medicine, 179, 108902, September 2024. doi:10.1016/j.compbiomed.2024.108902
[9] S. Shirae, S. S. Debsarkar, H. Kawanaka, B. J. Aronow, V. B. S. Prasath. Multimodal ensemble fusion deep learning using histopathological images and clinical data for glioma subtype classification. IEEE Access, 2025. doi:10.1109/ACCESS.2025.3556713
[10] S. Shirae, S. S. Debsarkar, H. Kawanaka, B. J. Aronow, V. B. S. Prasath. Anonymous Submission. Submitted, 2024.
[11] S. Shirae, S. S. Debsarkar, H. Kawanaka, B. J. Aronow, V. B. S. Prasath. Multiple instance learning using reduced deep learning model for glioma subtype classification. Submitted, 2025.
In preparation:
Explainable deep learning models for automatic disease stage prediction in glioma histopathology images.
MS Theses:
Satoshi Shirae. Ensemble deep learning model for glioma classification and subtype clustering with histopathological imaging and clinical data. Graduate School of Engineering, Mie University, Japan, 2024.
Riku Nakagaki. Deep learning-based IDH1 gene mutation prediction using histopathological imaging and clinical data. Graduate School of Engineering, Mie University, Japan, 2024.
Kiichi Fukuma. Feature extraction and disease stage classification for glioma histopathology images. Graduate School of Engineering, Mie University, Japan, 2016.
Asami Yonekura. A study on subtypes clustering of glioma image using cell nuclei features. Graduate School of Engineering, Mie University, Japan, 2016.
Back to Histopathology Projects Main Page. Back to Research Page.