Skip to main content
main-content
Top

Hint

Swipe to navigate through the chapters of this book

2021 | OriginalPaper | Chapter

Combining Attention-Based Multiple Instance Learning and Gaussian Processes for CT Hemorrhage Detection

Authors : Yunan Wu, Arne Schmidt, Enrique Hernández-Sánchez, Rafael Molina, Aggelos K. Katsaggelos

Published in: Medical Image Computing and Computer Assisted Intervention – MICCAI 2021

Publisher: Springer International Publishing

share
SHARE

Abstract

Intracranial hemorrhage (ICH) is a life-threatening emergency with high rates of mortality and morbidity. Rapid and accurate detection of ICH is crucial for patients to get a timely treatment. In order to achieve the automatic diagnosis of ICH, most deep learning models rely on huge amounts of slice labels for training. Unfortunately, the manual annotation of CT slices by radiologists is time-consuming and costly. To diagnose ICH, in this work, we propose to use an attention-based multiple instance learning (Att-MIL) approach implemented through the combination of an attention-based convolutional neural network (Att-CNN) and a variational Gaussian process for multiple instance learning (VGPMIL). Only labels at scan-level are necessary for training. Our method (a) trains the model using scan labels and assigns each slice with an attention weight, which can be used to provide slice-level predictions, and (b) uses the VGPMIL model based on low-dimensional features extracted by the Att-CNN to obtain improved predictions both at slice and scan levels. To analyze the performance of the proposed approach, our model has been trained on 1150 scans from an RSNA dataset and evaluated on 490 scans from an external CQ500 dataset. Our method outperforms other methods using the same scan-level training and is able to achieve comparable or even better results than other methods relying on slice-level annotations.
Literature
2.
go back to reference Arbabshirani, M.R., et al.: Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration. NPJ Digital Med. 1(1), 1–7 (2018) CrossRef Arbabshirani, M.R., et al.: Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration. NPJ Digital Med. 1(1), 1–7 (2018) CrossRef
3.
go back to reference Burduja, M., Ionescu, R.T., Verga, N.: Accurate and efficient intracranial hemorrhage detection and subtype classification in 3D CT scans with convolutional and long short-term memory neural networks. Sensors 20(19), 5611 (2020) CrossRef Burduja, M., Ionescu, R.T., Verga, N.: Accurate and efficient intracranial hemorrhage detection and subtype classification in 3D CT scans with convolutional and long short-term memory neural networks. Sensors 20(19), 5611 (2020) CrossRef
4.
go back to reference Campanella, G., Silva, V.W.K., Fuchs, T.J.: Terabyte-scale deep multiple instance learning for classification and localization in pathology (2018) Campanella, G., Silva, V.W.K., Fuchs, T.J.: Terabyte-scale deep multiple instance learning for classification and localization in pathology (2018)
5.
go back to reference Chang, P., et al.: Hybrid 3D/2D convolutional neural network for hemorrhage evaluation on head CT. AJNR Am. J. Neuroradiol. 39(9), 1609–1616 (2018) CrossRef Chang, P., et al.: Hybrid 3D/2D convolutional neural network for hemorrhage evaluation on head CT. AJNR Am. J. Neuroradiol. 39(9), 1609–1616 (2018) CrossRef
6.
go back to reference Chilamkurthy, S., et al.: Development and validation of deep learning algorithms for detection of critical findings in head CT scans (2018) Chilamkurthy, S., et al.: Development and validation of deep learning algorithms for detection of critical findings in head CT scans (2018)
7.
go back to reference Haußmann, M., Hamprecht, F., Kandemir, M.: Variational Bayesian multiple instance learning with gaussian processes. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 810–819 (2017) Haußmann, M., Hamprecht, F., Kandemir, M.: Variational Bayesian multiple instance learning with gaussian processes. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 810–819 (2017)
8.
go back to reference Ilse, M., Tomczak, J.M., Welling, M.: Attention-based deep multiple instance learning (2018) Ilse, M., Tomczak, J.M., Welling, M.: Attention-based deep multiple instance learning (2018)
9.
go back to reference Jnawali, K., Arbabshirani, M.R., Rao, N., Patel, A.A.: Deep 3D convolution neural network for CT brain hemorrhage classification. In: Mori, K., Petrick, N. (eds.) Medical Imaging 2018: Computer-Aided Diagnosis, p. 47. SPIE (2018) Jnawali, K., Arbabshirani, M.R., Rao, N., Patel, A.A.: Deep 3D convolution neural network for CT brain hemorrhage classification. In: Mori, K., Petrick, N. (eds.) Medical Imaging 2018: Computer-Aided Diagnosis, p. 47. SPIE (2018)
10.
go back to reference Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization (2017) Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization (2017)
11.
go back to reference Monteiro, M., et al.: Multiclass semantic segmentation and quantification of traumatic brain injury lesions on head CT using deep learning: an algorithm development and multicentre validation study. Lancet Digital Health 2(6), e314–e322 (2020) CrossRef Monteiro, M., et al.: Multiclass semantic segmentation and quantification of traumatic brain injury lesions on head CT using deep learning: an algorithm development and multicentre validation study. Lancet Digital Health 2(6), e314–e322 (2020) CrossRef
12.
go back to reference Nguyen, N.T., Tran, D.Q., Nguyen, N.T., Nguyen, H.Q.: A CNN-LSTM architecture for detection of intracranial hemorrhage on CT scans (2020) Nguyen, N.T., Tran, D.Q., Nguyen, N.T., Nguyen, H.Q.: A CNN-LSTM architecture for detection of intracranial hemorrhage on CT scans (2020)
13.
go back to reference Otite, F.O., Khandelwal, P., Malik, A.M., Chaturvedi, S., Sacco, R.L., Romano, J.G.: Ten-year temporal trends in medical complications after acute intracerebral hemorrhage in the united states. Stroke 48(3), 596–603 (2017) CrossRef Otite, F.O., Khandelwal, P., Malik, A.M., Chaturvedi, S., Sacco, R.L., Romano, J.G.: Ten-year temporal trends in medical complications after acute intracerebral hemorrhage in the united states. Stroke 48(3), 596–603 (2017) CrossRef
14.
go back to reference Patel, A., Leemput, S.C.v.d., Prokop, M., Ginneken, B.V., Manniesing, R.: Image level training and prediction: intracranial hemorrhage identification in 3D non-contrast CT. IEEE Access 7, 92355–92364 (2019) Patel, A., Leemput, S.C.v.d., Prokop, M., Ginneken, B.V., Manniesing, R.: Image level training and prediction: intracranial hemorrhage identification in 3D non-contrast CT. IEEE Access 7, 92355–92364 (2019)
15.
go back to reference Remedios, S., et al.: Extracting 2D weak labels from volume labels using multiple instance learning in CT hemorrhage detection. In: Medical Imaging 2020: Image Processing, vol. 11313, p. 113130F (2020) Remedios, S., et al.: Extracting 2D weak labels from volume labels using multiple instance learning in CT hemorrhage detection. In: Medical Imaging 2020: Image Processing, vol. 11313, p. 113130F (2020)
17.
go back to reference Sato, D., et al.: A primitive study on unsupervised anomaly detection with an autoencoder in emergency head CT volumes. In: Medical Imaging 2018: Computer-Aided Diagnosis, p. 60 (2018) Sato, D., et al.: A primitive study on unsupervised anomaly detection with an autoencoder in emergency head CT volumes. In: Medical Imaging 2018: Computer-Aided Diagnosis, p. 60 (2018)
18.
go back to reference Sobrino, J., Shafi, S.: Timing and causes of death after injuries, vol. 26, no. 2, pp. 120–123 (2013) Sobrino, J., Shafi, S.: Timing and causes of death after injuries, vol. 26, no. 2, pp. 120–123 (2013)
19.
go back to reference Titano, J.J., et al.: Automated deep-neural-network surveillance of cranial images for acute neurologic events. Nat. Med. 24(9), 1337–1341 (2018) CrossRef Titano, J.J., et al.: Automated deep-neural-network surveillance of cranial images for acute neurologic events. Nat. Med. 24(9), 1337–1341 (2018) CrossRef
Metadata
Title
Combining Attention-Based Multiple Instance Learning and Gaussian Processes for CT Hemorrhage Detection
Authors
Yunan Wu
Arne Schmidt
Enrique Hernández-Sánchez
Rafael Molina
Aggelos K. Katsaggelos
Copyright Year
2021
DOI
https://doi.org/10.1007/978-3-030-87196-3_54

Premium Partner