ABSTRACT
Intracranial Hemorrhage (ICH), a dangerous and devastating medical emergency, affects thousands of patients every year around the world. In the clinical settings, Computer Tomography (CT), is widely used for diagnosis of neurological diseases. In the situation of Intracranial Hemorrhage, not only saving time is critically important, but also the expertise to accurately diagnose and locate ICH is imperative. However, there are not always enough doctors working in the emergency expert in the field of ICH, and the results from using only deep learning models are not always reliable.
Three neural networks, VGG-19, Resnet-101, and DenseNet-201 were trained separately on preprocessed the Intracranial hemorrhage data with labels and used the Grad-CAM method to produce a saliency map by visualizing the process of the network making a decision regarding to specific class index, thus increasing the interpretability of the results. We tested the networks' performances on our preprocessed CT data, and their differences produced saliency maps.
Three experiments were designed and conducted to help us understand our models' performance and predictions in different contexts. First, we observed the differences between the pre-trained deep learning model and the unpre-trained deep learning models. Second, we observed how the performance and Grad-CAM results would differ when the images were normalized at different Window values. Third, we merged the six Grad-CAM images generated by the six class indices for each image into a single image and fed it into the network to observe the results.
To further demonstrate the potential application of our deep learning models, we used trained models to make a GUI software called ICH Deep Learning Detector in python with the PyQt5 library to simplify the process of doctors using the deep learning model and learning from predictions.
- Caceres, J. Alfredo, and Joshua N. Goldstein. "Intracranial Hemorrhage." Emergency Medicine Clinics of North America, vol. 30, no. 3, Aug. 2012, pp. 771--794, www.ncbi.nlm.nih.gov/pmc/articles/PMC3443867/10.1016/j.emc.2012.06.003.Google ScholarCross Ref
- Cordonnier, Charlotte, et al. "Intracerebral Haemorrhage: Current Approaches to Acute Management." The Lancet, vol. 392, no. 10154, Oct. 2018, pp. 1257--1268, www.thelancet.com/article/S0140-6736(18)31878-6/fulltext, 10.1016/s0140-6736(18)31878-6. Accessed 30 Oct. 2019.Google ScholarCross Ref
- Elliott J, Smith M. The Acute Management of Intracerebral Hemorrhage: A Clinical Review. Anesthesia & Analgesia. 2010;110(5):1419--1427.Google Scholar
- Panagos, Peter D., et al. "Intracerebral Hemorrhage." Emergency Medicine Clinics of North America, vol. 20, no. 3, Aug. 2002, pp. 631--655, 10.1016/s0733-8627(02)00015-9.Google ScholarCross Ref
- Mitchell, Thomas M. Machine Learning. Singapore, Mcgraw-Hill, 1997.Google ScholarDigital Library
- Flanders, Adam E., et al. "Construction of a Machine Learning Dataset through Collaboration: The RSNA 2019 Brain CT Hemorrhage Challenge." Radiology: Artificial Intelligence, vol. 2, no. 3, 1 May 2020, p. e190211, 10.1148/ryai.2020190211. Accessed 5 Sept. 2020.Google Scholar
- Awwal Muhammad Dawud, Kamil Yurtkan, Huseyin Oztoprak, "Application of Deep Learning in Neuroradiology: Brain Haemorrhage Classification Using Transfer Learning", Computational Intelligence and Neuroscience, vol. 2019, Article ID 4629859, 12 pages, 2019. https://doi.org/10.1155/2019/4629859Google ScholarCross Ref
- Majumdar A, Brattain L, Telfer B, Farris C, Scalera J. Detecting Intracranial Hemorrhage with Deep Learning. Conf Proc IEEE Eng Med Biol Soc 2018; 2018:583--587.Google Scholar
- Li, Q., Cai, W., Wang, X., Zhou, Y., Feng, D. D., & Chen, M. (1997). Medical image classification with convolutional neural network. In 2014 13th International Conference on Control Automation Robotics and Vision, ICARCV 2014 (pp. 844--848). [7064414] Institute of Electrical and Electronics Engineers Inc. https://doi.org/10.1109/ICARCV.2014.7064414Google ScholarCross Ref
- M. Kallenberg et al., "Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammographic Risk Scoring," in IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1322--1331, May 2016, doi: 10.1109/TMI.2016.2532122.Google ScholarCross Ref
- Kuo W, Häne C, Mukherjee P, Malik J, Yuh EL. Expert-level detection of acute intracranial hemorrhage on head computed tomography using deep learning. Proc Natl Acad Sci U S A 2019;116(45):22737--22745.Google Scholar
- Dawud AM, Yurtkan K, Oztoprak H. Application of Deep Learning in Neuroradiology: Brain Haemorrhage Classification Using Transfer Learning. Comput Intell Neurosci 2019; 2019:4629859.Google Scholar
- Wikipedia Contributors. "Python (Programming Language)." Wikipedia, Wikimedia Foundation, 4 May 2019, en.wikipedia.org/wiki/Python_(programming_language).Google Scholar
- "RSNA Intracranial Hemorrhage Detection." Kaggle.Com, www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/data. Accessed 5 Sept. 2020.Google Scholar
- B. Zhou, A. Khosla, L. A., A. Oliva, and A. Torralba. Learning Deep Features for Discriminative Localization. In CVPR, 2016. 2, 3, 5, 6, 20Google ScholarCross Ref
- "Pydicom Documentation --- Pydicom 1.0a Documentation." No-Ip.Org, 2012, lira.no-ip.org:8080/doc/python-dicom-doc/html/. Accessed 11 Sept. 2020.Google Scholar
- Kingma, Diederik P, and Jimmy Ba. "Adam: A Method for Stochastic Optimization." ArXiv.Org, 2014, arxiv.org/abs/1412.6980.Google Scholar
- Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2020). Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization. International Journal of Computer Vision, 128(2), 336-- 359. https://doi.org/10.1007/s11263-019-01228-7Google ScholarDigital Library
- CrossEntropyLoss --- PyTorch 1.6.0 documentation. (n.d.). Pytorch.Org. Retrieved September 11, 2020, from https://pytorch.org/docs/stable/generated/torch.nn.CrossEntropyLoss.htmlGoogle Scholar
- Simonyan, K., & Zisserman, A. (2014). Very Deep Convolutional Networks for Large-Scale Image Recognition. ArXiv.Org. https://arxiv.org/abs/1409.1556Google Scholar
- Tsang, S.-H. (2019, March 20). Review: ResNet --- Winner of ILSVRC 2015 (Image Classification, Localization, Detection). Medium. https://towardsdatascience.com/review-resnet-winner-of-ilsvrc-2015-image-classification-localization-detection-e39402bfa5d8Google Scholar
- Huang, G., Liu, Z., van, & Weinberger, Kilian Q. (2016). Densely Connected Convolutional Networks. ArXiv.Org. https://arxiv.org/abs/1608.06993Google Scholar
- CNN Architectures: VGG, ResNet, Inception + TL. (n.d.). Kaggle.Com. Retrieved September 12, 2020, from https://www.kaggle.com/shivamb/cnn-architectures-vgg-resnet-inception-tlGoogle Scholar
- Bonner, A. (2019, June 1). The Complete Beginner's Guide to Deep Learning: Convolutional Neural Networks. Medium. https://towardsdatascience.com/wtf-is-image-classification8e78a8235acb#:~:text=The%20convolutional%20neural%20network%20(CNNGoogle Scholar
- Digital Imaging and Communications in Medicine - an overview | ScienceDirect Topics. (n.d.). Www.Sciencedirect.Com. https://www.sciencedirect.com/topics/biochemistry-genetics-and-molecular-biology/digital-imaging-and-communications-in-medicineGoogle Scholar
Index Terms
- An Interpretable Deep Learning System for Automatic Intracranial Hemorrhage Diagnosis with CT Image
Recommendations
Automated Detection of Intracranial Hemorrhage on Head Computed Tomography with Deep Learning
ICBET '20: Proceedings of the 2020 10th International Conference on Biomedical Engineering and TechnologyIntracranial hemorrhage is a serious health problem worldwide requiring rapid and often intensive medical treatment. However, the diagnosis process of intracranial hemorrhage is complicated and often time consuming when looking for the presence, ...
Siamese U-Net with Healthy Template for Accurate Segmentation of Intracranial Hemorrhage
Medical Image Computing and Computer Assisted Intervention – MICCAI 2019AbstractIntracranial hemorrhage (ICH) is a fatal form of stroke which is caused by bleeding within or around the brain. Detection and quantification of hemorrhage are critical in the diagnosis and treatment of the disease. In this paper, we propose ...
Deep Learning Models for Intracranial Hemorrhage Recognition: A comparative study
AbstractEvery day, a large number of people with brain injury are received in the emergency rooms. Due to the large number of slices analyzed by the doctors for each patient and to accelerate the diagnosis, the development of a precise computer-aided ...
Comments