Explainable AI for Transparent MRI Segmentation: Deep Learning and Visual Attribution in Clinical Decision Support
DOI:
https://doi.org/10.22399/ijcesen.479Keywords:
Explainable AI, MRI, segmentationAbstract
For medical diagnosis and therapy planning, the importance of accurate MRI segmentation cannot be overemphasized. Conversely, the inscrutability of deep learning models remains obstacles to their application in therapeutic contexts. In this article, an interpretability artificial intelligence framework is introduced. It combines an MRI segmentation model based on deep learning, visual attribution algorithms and natural language explanations. EXPERIMENT The dataset is consisting of plenty of different types of brain MRI scans, and used to test the architecture. The average of Dice score of our method is 88.7% and 92.3% for segmentation of tumor and categorization of tissues, respectively. Both are pretty epic scores. The insights extracted from both the visual attributions and textual explations improve our understanding of how the model arrives at its decisions, thereby increasing the transparency and interpretability of the model. believe this approach to explainable artificial intelligence can help to close the gap between state-of-the-art performance in MRI segmentation and clinical interpretability, by increasing the transparency of complex models and facilitating their implementation into a clinical workflow. Conclusion Our approach may have implications in the transparent and reliable development of AI-based decision support systems for medical imaging
References
I. Ahmad, Zeeshan Asghar, Tanesh Kumar, Gaolei Li, Ahsan Manzoor et al., (2022). Emerging Technologies for Next Generation Remote Health Care and Assisted Living. in IEEE Access,10:56094-56132
Mostafa AM, Zakariah M, Aldakheel EA. (2023). Brain Tumor Segmentation Using Deep Learning on MRI Images. Diagnostics.13(9):15.
Mostafa, A.M.,Zakariah, M., Aldakheel, E.A. (2023) Brain Tumor Segmentation Using Deep Learning on MRI Images. Diagnostics 13;1562.
Ozkara BB, Chen MM, Federau C, Karabacak M, Briere TM, Li J, Wintermark M. (2023). Deep Learning for Detecting Brain Metastases on MRI: A Systematic Review and Meta-Analysis. Cancers (Basel). 15(2):334. doi: 10.3390/cancers15020334. PMID: 36672286; PMCID: PMC9857123.
Ma, K.; He, S.; Sinha, G.; Ebadi, A.; Florea, A.; Tremblay, S.; Wong, A.; Xi, P. (2023). Towards Building a Trustworthy Deep Learning Framework for Medical Image Analysis. Sensors 23,8122. https://doi.org/10.3390/s23198122
Tang X. (2019). The role of artificial intelligence in medical imaging research. BJR Open. 2(1):20190031.
Hoadley KA, Yau C, Lawrence MS, Noushmehr H, Malta TM et al., (2018). Cancer Genome Atlas Network; Stuart JM, Benz CC, Laird PW. Cell-of-Origin Patterns Dominate the Molecular Classification of 10,000 Tumors from 33 Types of Cancer. Cell. 173(2):291-304.
Katarzyna Borys, Yasmin Alyssa Schmitt, Meike Nauta, Christin Seifert, Nicole Krämer, Christoph M. Friedrich, Felix Nensa, (2023). Explainable AI in medical imaging: An overview for clinical practitioners – Beyond saliency-based XAI approaches, European Journal of Radiology, 162,110786,
D. Cheng, Mengying Xiao, Liyuan Zhang et al., (2023). Visually explaining medical image diagnosis using Grad-CAM: A review. Biomedical Signal Processing and Control, 80;104263
G. R. Wu, M. Kim, Q. Wang, Y. Z. Gao, S. Liao, and D. G. Shen, (2013). Unsupervised deep feature learning for deformable registration of MR brain images, in Proc. 16th Int. Conf. Medical Image Computing and Computer-Assisted Intervention, Nagoya, Japan, pp. 649–656
Shen SY, Singhania R, Fehringer G, Chakravarthy A, Roehrl MHA (2018). Sensitive tumour detection and classification using plasma cell-free DNA methylomes. Nature. 563(7732):579-583.
Jin Liu, Yi Pan, Min Li, Ziyue Chen., A. Garcia et al., (2018). Big data mining and analytics 1(1);1– 18.,
O. Ronneberger, P. Fischer, and T. Brox, (2023). U-Net: Convolutional networks for biomedical image segmentation. MICCAI 2015, pp. 234-241.
J. Schlemper., Ashish Sinha et al (2023). Attention gated networks: Learning to leverage salient regions in medical images. Medical Image Analysis, 53;197-207.
X. Li, H. Chen, X. Qi, Q. Dou, C. -W. Fu and P. -A. Heng, (2018). H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes. IEEE Transactions on Medical Imaging, 37(12);2663-2674
Y. Xue et al., Cheng Chen, Siyu Qi, Kangneng Zhou, Tong Lu, Huansheng Ning and Ruoxiu Xiao (2023). SegAN: Adversarial network with multi-scale L1 loss for medical image segmentation. Neuroinformatics, 18;1-18.
Zhang, A., Xing, L., Zou, J. et al. (2022). Shifting machine learning for healthcare from development to deployment and from models to data. Nat. Biomed. Eng 6;1330–1345.
R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, (2023). Grad-CAM: Visual explanations from deep networks via gradient-based localization," ICCV 2017, pp. 618-626.
S. Bach, A. Binder, G. Montavon, F. Klauschen, K.-R. Müller, and W. Samek, (2023). On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE, 10(7);e0130140.
Tran N., (2023). Lauw H Memory Network-Based Interpreter of User Preferences in Content-Aware Recommender Systems ACM Transactions on Intelligent Systems and Technology 14(6);1-28 DOI: 10.1145/3625239
T. Panigutti et al., (2023). Doctor XAI: An ontology-based approach to black-box sequential data classification explanations. ACM Conference on Fairness, Accountability, and Transparency (FAccT) 2020, pp. 629-639.
Shensi Shen, Stéphan Vagner, Caroline Robert, (2023). An explainable deep learning framework for brain tumor segmentation and visual interpretation," IEEE 9;54998-55008.
Y. Wang e., Risheng Wang, Tao Lei, Ruixia Cui., (2023). Explainable medical image segmentation with joint learning of segmentation and explanation, IEEE Transactions on Medical Imaging, 41(7);1749-1760.
H. Guo et al., Bingtao Zhang, Hongying Meng, Asoke K. Nandi (2023). BrainExplainer: An explainable AI framework for brain tumor segmentation and explanation generation," Medical Image Analysis, 80;102529.
Dhar, T., Dey, N., Borra, S. and Sherratt, R. S, (2023). Challenges of Deep Learning in Medical Image Analysis -Improving Explainability and Trust, IEEE Transactions on Technology and Society PP(99).
F. Milletari., Stefan Bauer., Jayashree Kalpathy., Cramer et al., (2023). V-Net: Fully convolutional neural networks for volumetric medical image segmentation," 3DV 2016, pp. 565-571.
Sepp Hochreiter, Jurgen Schmidhuber., (1997). Long Short-Term Memory. Neural Comput 9(8): 1735–1780.
Bjoern H. Menze, Andras Jakab et al., ()"The multimodal brain tumor image segmentation benchmark (BRATS)," IEEE Transactions on Medical Imaging, 34(10);1993-2024.
Hernandez Petzsche, M.R., de la Rosa, E., Hanning, U. et al., (2023). ISLES 2015 - A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI. Medical Image Analysis, 35;250-269.
D. P. Huttenlocher., Manuel Bogoya., Vargas., et al., (2023). Comparing images using the Hausdorff distance IEEE Transactions on Pattern Analysis and Machine Intelligence, 15(9);850-863.
K. Papineni, S. Roukos, T. Ward, and W.-J. Zhu, (2023). BLEU: A method for automatic evaluation of machine translation. ACL 2002, pp. 311-318.
C. Y. Lin, (2023). ROUGE: A package for automatic evaluation of summaries. ACL Workshop on Text Summarization Branches Out, pp. 74-81.
Bakas, S., Akbari, H., Sotiras, A. et al., (2023). Advancing the Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features. Scientific Data, 4;170117.
Oskar Maier, Bjoern H Menze , Janina von der Gablentz , Levin Han., et al., (2023). ISLES 2015-2017: A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI. Medical Image Analysis, 67;101849.
S. M. Smith, (2023). Fast robust automated brain extraction. Human Brain Mapping, 17(3);143-155.
Ullah F, Nadeem M, Abrar M, Al-Razgan M, Alfakih T, Amin F, Salam A. (2023). Brain Tumor Segmentation from MRI Images Using Handcrafted Convolutional Neural Network. Diagnostics (Basel).13(16):2650.
P. Thevenaz et al., (2023). Image interpolation and resampling. Handbook of Medical Image Processing and Analysis, pp. 465-493, 2023.
Olaf Ronneberger, Philipp Fischer, Thomas Brox, (2023). U-Net: Convolutional networks for biomedical image segmentation," MICCAI 2015, pp. 234-241.
O. Oktay., Jo Schlemper, Loic Le Folgoc, Matthew Lee et al., (2023). Attention U-Net: Learning where to look for the pancreas. arXiv preprint arXiv:1804.03 pp 999.
F. Milletari, N. Navab, and S.-A. Ahmadi, (2023). V-Net: Fully convolutional neural networks for volumetric medical image segmentation. 3DV 2016, pp. 565-571.
C. H. Sudre et al., (2023). Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations," DLMIA 2017, pp. 240-248.
R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. ICCV 2017, pp. 618-626.
S. Bach, A. Binder, G. Montavon, F. Klauschen, K.-R. Müller, and W. Samek, (2023). On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation," PLoS ONE, 10(7);e0130140.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
