Next Article in Journal
UWB-Based Self-Localization Strategies: A Novel ICP-Based Method and a Comparative Assessment for Noisy-Ranges-Prone Environments
Next Article in Special Issue
Depth Estimation for Light-Field Images Using Stereo Matching and Convolutional Neural Networks
Previous Article in Journal
An Adaptive Multi-Channel Cooperative Data Transmission Scheduling in VANETs
Previous Article in Special Issue
3DMGNet: 3D Model Generation Network Based on Multi-Modal Data Constraints and Multi-Level Feature Fusion
Article

Accurate and Efficient Intracranial Hemorrhage Detection and Subtype Classification in 3D CT Scans with Convolutional and Long Short-Term Memory Neural Networks

1
Department of Computer Science, University of Bucharest, 14 Academiei, 010014 Bucharest, Romania
2
Romanian Young Academy, University of Bucharest, 90 Panduri, 050663 Bucharest, Romania
3
Department of Radiotherapy, Oncology and Hematology, “Carol Davila” University of Medicine and Pharmacy, 27 Dionisie Lupu, 020021 Bucharest, Romania
*
Authors to whom correspondence should be addressed.
Sensors 2020, 20(19), 5611; https://doi.org/10.3390/s20195611
Received: 1 August 2020 / Revised: 18 September 2020 / Accepted: 27 September 2020 / Published: 1 October 2020
(This article belongs to the Special Issue Sensors and Computer Vision Techniques for 3D Object Modeling)
In this paper, we present our system for the RSNA Intracranial Hemorrhage Detection challenge, which is based on the RSNA 2019 Brain CT Hemorrhage dataset. The proposed system is based on a lightweight deep neural network architecture composed of a convolutional neural network (CNN) that takes as input individual CT slices, and a Long Short-Term Memory (LSTM) network that takes as input multiple feature embeddings provided by the CNN. For efficient processing, we consider various feature selection methods to produce a subset of useful CNN features for the LSTM. Furthermore, we reduce the CT slices by a factor of 2×, which enables us to train the model faster. Even if our model is designed to balance speed and accuracy, we report a weighted mean log loss of 0.04989 on the final test set, which places us in the top 30 ranking (2%) from a total of 1345 participants. While our computing infrastructure does not allow it, processing CT slices at their original scale is likely to improve performance. In order to enable others to reproduce our results, we provide our code as open source. After the challenge, we conducted a subjective intracranial hemorrhage detection assessment by radiologists, indicating that the performance of our deep model is on par with that of doctors specialized in reading CT scans. Another contribution of our work is to integrate Grad-CAM visualizations in our system, providing useful explanations for its predictions. We therefore consider our system as a viable option when a fast diagnosis or a second opinion on intracranial hemorrhage detection are needed. View Full-Text
Keywords: intracranial hemorrhage detection; intracranial hemorrhage subtype classification; convolutional neural networks; long short-term memory networks intracranial hemorrhage detection; intracranial hemorrhage subtype classification; convolutional neural networks; long short-term memory networks
Show Figures

Figure 1

MDPI and ACS Style

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 2020, 20, 5611. https://doi.org/10.3390/s20195611

AMA Style

Burduja M, Ionescu RT, 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. 2020; 20(19):5611. https://doi.org/10.3390/s20195611

Chicago/Turabian Style

Burduja, Mihail, Radu T. Ionescu, and Nicolae Verga. 2020. "Accurate and Efficient Intracranial Hemorrhage Detection and Subtype Classification in 3D CT Scans with Convolutional and Long Short-Term Memory Neural Networks" Sensors 20, no. 19: 5611. https://doi.org/10.3390/s20195611

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop