Ensembled Deep Neural Network for Intracranial Hemorrhage Detection and Subtype Classification on Noncontrast CT Images


  • Yunan Wu Department of Diagnostic Radiology, Rush University Medical Center, 1653 W. Congress Pkwy, Jelke Street 181, Chicago, IL, 60612, USA https://orcid.org/0000-0001-6980-9746
  • Mark P. Supanich 1Department of Diagnostic Radiology, Rush University Medical Center, 1653 W. Congress Pkwy, Jelke Street 181, Chicago, IL, 60612, USA https://orcid.org/0000-0001-9514-9390
  • Deng Jie Department of Diagnostic Radiology, Rush University Medical Center, 1653 W. Congress Pkwy, Jelke Street 181, Chicago, IL, 60612, USA




Intracranial hemorrhage, Subtype classification, Computer tomography, Deep learning, Ensembled model


Rapid and accurate diagnosis of intracranial hemorrhage is clinically significant to ensure timely treatment. In this study, we developed an ensembled deep neural network for the detection and subtype classification of intracranial hemorrhage. The model consisted of two parallel network pathways, one using three different window level/width settings to enhance the image contrast of brain, blood, and soft tissue. The other extracted spatial information of adjacent image slices to the target slice. Both pathways exploited the EfficientNet-B0 as the basic architecture and were ensembled to generate the final prediction. Class activation mapping was applied in both pathways to highlight the regions of detected hemorrhage and the associated subtypes. The model was trained and tested using Intracranial Hemorrhage Detection Challenge (IHDC) dataset launched by the Radiological Society of North America (RSNA) in 2019, which contained 674,258 head noncontrasts computer tomography images acquired from 19,530 patients. An independent dataset (CQ500) acquired from another institution was used to test the generalizability of the trained model. The overall accuracy, sensitivity, and F1 score for intracranial hemorrhage detection were 95.7%, 85.9%, and 86.7% on IHDC testing dataset and 92.4%, 92.6%, and 93.4% on external CQ500 testing dataset. The heatmaps by class activation mapping successfully demonstrated discriminative feature regions of the predicted hemorrhage locations and subtypes, providing visual guidance for radiologists to assist in rapid diagnosis of intracranial hemorrhage.


F.O. Otite, P. Khandelwal, A.M. Malik, S. Chaturvedi, R.L. Sacco, J.G. Romano, Ten-year temporal trends in medical complications after acute intracerebral hemorrhage in the United States, Stroke. 48 (2017), 596–603.

I. Hostettler, D. Seiffge, D. Werring, Intracerebral hemorrhage: an update on diagnosis and treatment, Expert Rev. Neurother. 19 (2019), 679–694.

W.M. Strub, J.L. Leach, T. Tomsick, A. Vagal, Overnight preliminary head CT interpretations provided by residents: locations of misidentified intracranial hemorrhage, Am. J. Neuroradiol. 28 (2007), 1679–1682.

D. Alfaro, M.A. Levitt, D.K. English, V. Williams, R. Eisenberg, Accuracy of interpretation of cranial computed tomography scans in an emergency medicine residency program, Ann. Emerg. Med. 25 (1995), 169–174.

W.K. Erly, W.G. Berger, E. Krupinski, J.F. Seeger, J.A. Guisto, Radiology resident evaluation of head CT scan orders in the emergency department, AJNR Am. J. Neuroradiol. 23 (2002), 103–107.

N.R. Lal, U.M. Murray, O.P. Eldevik, J.S. Desmond, Clinical consequences of misinterpretations of neuroradiologic CT scans by on-call radiology residents, AJNR Am. J. Neuroradiol. 21 (2000), 124–129.

J.K. Wasserman, L.C. Schlichter, Neuron death and inflammation in a rat model of intracerebral hemorrhage: effects of delayed minocycline treatment, Brain Res. 1136 (2007), 208–218.

F. Siddiq, S.A. Chaudhry, R.P. Tummala, M.F.K. Suri, A.I. Qureshi, Factors and outcomes associated with early and delayed aneurysm treatment in subarachnoid hemorrhage patients in the United States, Neurosurgery. 71 (2012), 670–678.

K.N. Keshavamurthy, O.P. Leary, L.H. Merck, B. Kimia, S. Collins, D.W. Wright, et al., Machine learning algorithm for automatic detection of CT-identifiable hyperdense lesions associated with traumatic brain injury, in Medical Imaging 2017: ComputerAided Diagnosis, International Society for Optics and Photonics, Orlando, FL, USA, 2017, p. 101342G.

L.A. Ramos, W.E. van der Steen, R. Sales Barros, C.B.L.M. Majoie, R. van den Berg, D. Verbaan, et al., Machine learning improves prediction of delayed cerebral ischemia in patients with subarachnoid hemorrhage, J. NeuroInterventional Surg. 11 (2019), 497–502.

T.D. Phong, H.N. Duong, H.T. Nguyen, N.T. Trong, V.H. Nguyen, T. Van Hoa, et al., Brain hemorrhage diagnosis by using deep learning, in Proceedings of the 2017 International Conference on Machine Learning and Soft Computing, Association for Computing Machinery, New York, NY, USA, 2017, pp. 34–39.

M. Karki, J. Cho, E. Lee, M.-H. Hahm, S.-Y. Yoon, M. Kim, et al., CT window trainable neural network for improving intracranial hemorrhage detection by combining multiple settings, Artif. Intell. Med. 106 (2020), 101850.

A. Sage, P. Badura, Intracranial hemorrhage detection in head CT using double-branch convolutional neural network, support vector machine, and random forest, Appl. Sci. 10 (2020), 7577.

M. Grewal, M.M. Srivastava, P. Kumar, S. Varadarajan, RADnet: radiologist level accuracy using deep learning for hemorrhage detection in CT scans, in 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC, USA, 2018, pp. 281–284.

J. Ker, S.P. Singh, Y. Bai, J. Rao, T. Lim, L. Wang, Image thresholding improves 3-dimensional convolutional neural network diagnosis of different acute brain hemorrhages on computed tomography scans, Sensors. 19 (2019), 2167.

W. Kuo, C. Hne, P. Mukherjee, J. Malik, E.L. Yuh, Expert-level detection of acute intracranial hemorrhage on head computed tomography using deep learning, Proc. Natl. Acad. Sci. 116 (2019), 22737–22745.

M. Burduja, R.T. Ionescu, N. Verga, Accurate and efficient intracranial hemorrhage detection and subtype classification in 3D CT scans with convolutional and long short-term memory neural networks, Sensors. 20 (2020), 5611.

N.T. Nguyen, D.Q. Tran, N.T. Nguyen, H.Q. Nguyen, A CNNLSTM architecture for detection of intracranial hemorrhage on CT scans, ArXiv200510992 Cs, 2020.

B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, A. Torralba, Learning deep features for discriminative localization, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Las Vegas, NV, USA, 2016, pp. 2921–2929.

RSNA, Intracranial Hemorrhage Detection Challenge, 2019. https://www.kaggle.com/c/rsna-intracranial-hemorrhagedetection.

S. Chilamkurthy, R. Ghosh, S. Tanamala, M. Biviji, N.G. Campeau, V.K. Venugopal, et al., Development and validation of deep learning algorithms for detection of critical findings in head CT scans, ArXiv180305854 Cs, 2018.

Z. Xue, S. Antani, L.R. Long, D. Demner-Fushman, G.R. Thoma, Window classification of brain CT images in biomedical articles, AMIA Ann. Symp. Proc. 2012 (2012),1023–1029.

M. Tan, Q.V. Le, EfficientNet: rethinking model scaling for convolutional neural networks, ArXiv190511946 Cs Stat, 2020.

N.K. Chowdhury, M.A. Kabir, M.M. Rahman, N. Rezoana, ECOVNet: an ensemble of deep convolutional neural networks based on EfficientNet to detect COVID-19 from chest X-rays, ArXiv200911850 Cs Eess, 2020.

G. Marques,D. Agarwal, I. de la Torre Díez, Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network, App. Soft Comput. 96 (2020), 106691.

H. Wu, H. Yin, H. Chen, M. Sun, X. Liu, Y. Yu, et al., A deep learning, image based approach for automated diagnosis for inflammatory skin diseases, Ann. Transl. Med. 8 (2020), 581.

M. Sandler,A. Howard, M. Zhu, A. Zhmoginov, L.-C. Chen, MobileNetV2: inverted residuals and linear bottlenecks, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 4510–4520.

J. Hu, L. Shen, S. Albanie, G. Sun, E. Wu, Squeeze-and-excitation networks, ArXiv170901507 Cs, 2019.

J. Deng, W. Dong, R. Socher, L. Li, K. Li, L. Fei-Fei, ImageNet: a large-scale hierarchical image database, in 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, 2009, pp. 248–255.

D.P. Kingma, J. Ba, Adam: a method for stochastic optimization, ArXiv14126980 Cs, 2017.

G. Forman, M. Scholz, Apples-to-apples in cross-validation studies: pitfalls in classifier performance measurement, ACM SIGKDD Explor Newsl. 12 (2010), 49–57.

Y. Cao, T.A. Geddes, J.Y.H. Yang, P. Yang, Ensemble deep learning in bioinformatics, Nat. Mach. Intell. 2 (2020), 500–508.

T.B. Chandra, K. Verma, B.K. Singh, D. Jain, S.S. Netam, Coronavirus disease (COVID-19) detection in chest X-ray images using majority voting based classifier ensemble, Expert Syst. Appl. 165 (2021), 113909.

H. Lee, S. Yune, M. Mansouri, M. Kim, S.H. Tajmir, C.E. Guerrier, et al., An explainable deep-learning algorithm for the detection of acute intracranial haemorrhage from small datasets, Nat. Biomed. Eng. 3 (2019), 173–182.

G. Danilov, K. Kotik, A. Negreeva, T. Tsukanova, M. Shifrin, N. Zakharova, et al., Classification of intracranial hemorrhage subtypes using deep learning on CT scans, Stud. Health Technol. Inform. 272 (2020), 370–373.

J. Zhang, B. Kailkhura, T.Y.-J. Han, Mix-n-match: ensemble and compositional methods for uncertainty calibration in deep learning, ArXiv200307329 Cs Stat, 2020.



How to Cite

Wu Y, Supanich MP, Jie D. Ensembled Deep Neural Network for Intracranial Hemorrhage Detection and Subtype Classification on Noncontrast CT Images. JAIMS [Internet]. 2021 May 5 [cited 2023 Feb. 4];2(1-2):12-20. Available from: http://oapublishing-jaims.com/jaims/article/view/62