Abstract

Deep learning methodologies constitute nowadays the main approach for medical image analysis and disease prediction. Large annotated databases are necessary for developing these methodologies; such databases are difficult to obtain and to make publicly available for use by researchers and medical experts. In this work, we focus on diagnosis of Covid-19 and its severity based on chest 3-D CT scans and develop a dual knowledge framework, including a large imaging database and a novel deep neural architecture. We introduce COV19-CT-DB, a very large database annotated for COVID-19 (detection and severity) that consists of 6,600 3-D CT scans (1,550 COVID-19 cases and 5,050 non-COVID-19 ones) of around 2,500,000 CT slices of more than 1,150 patients and 2,600 non-patients.
Harmonizing the analysis of data, especially of 3-D image volumes, consisting of different number of slices and annotated per volume, is a significant problem in training and using deep neural networks in various applications, including medical imaging. Moreover, unifying the decision making of the networks over different input datasets is crucial for the generation of rich data-driven knowledge and for trusted usage in the applications. This work further presents a new deep neural architecture, named RACNet, which includes routing and feature alignment steps and effectively handles different input lengths and single annotations of the 3-D image inputs, whilst providing highly accurate decisions. In addition, through latent variable extraction from the trained RACNet, a set of anchors are generated providing further insight on the network’s decision making. These can be used to enrich and unify data-driven knowledge extracted from different datasets. 

How to acquire the data

If you want access to the dataset, email me at: d.kollias@qmul.ac.uk from your official academic email (as data cannot be released to personal emails) with subject: COV19-CT-DB dataset request and include your job/title and official academic website in the email body.

References:

If you use the above data, you must cite the following papers:
 

• D. Kollias, et. al.: “A deep neural architecture for harmonizing 3-D input data analysis and decision making in medical imaging“, Neurocomputing, 2023.

@article{kollias2023deep, title={A deep neural architecture for harmonizing 3-D input data analysis and decision making in medical imaging}, author={Kollias, Dimitrios and Arsenos, Anastasios and Kollias, Stefanos}, journal={Neurocomputing}, volume={542}, pages={126244}, year={2023}, publisher={Elsevier}}

• D. Kollias, et. al.: “AI-MIA: COVID-19 Detection & Severity Analysis through Medical Imaging“, ECCV, 2022.

@inproceedings{kollias2022ai, title={Ai-mia: Covid-19 detection and severity analysis through medical imaging}, author={Kollias, Dimitrios and Arsenos, Anastasios and Kollias, Stefanos}, booktitle={European Conference on Computer Vision}, pages={677--690}, year={2022}, organization={Springer}}

• A. Arsenos, et. al.: “A Large Imaging Database and Novel Deep Neural Architecture for Covid-19 Diagnosis“, IVMSP, 2022.

@inproceedings{arsenos2022large, title={A large imaging database and novel deep neural architecture for Covid-19 diagnosis}, author={Arsenos, Anastasios and Kollias, Dimitrios and Kollias, Stefanos}, booktitle={2022 IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)}, pages={1--5}, year={2022}, organization={IEEE}}

• D. Kollias, et. al.: “MIA-COV19D: COVID-19 Detection through 3-D Chest CT Image Analysis“, ICCV, 2021.

@inproceedings{kollias2021mia, title={Mia-cov19d: Covid-19 detection through 3-d chest ct image analysis}, author={Kollias, Dimitrios and Arsenos, Anastasios and Soukissian, Levon and Kollias, Stefanos}, booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision}, pages={537–544}, year={2021} }

• D. Kollias, et. al.: “Deep transparent prediction through latent representation analysis“, 2020.

@article{kollias2020deep, title={Deep transparent prediction through latent representation analysis}, author={Kollias, Dimitrios and Bouas, N and Vlaxos, Y and Brillakis, V and Seferis, M and Kollia, Ilianna and Sukissian, Levon and Wingate, James and Kollias, S}, journal={arXiv preprint arXiv:2009.07044}, year={2020}}

• D. Kollias, et. al.: “Transparent Adaptation in Deep Medical Image Diagnosis”, TAILOR, 2020

@inproceedings{kollias2020transparent, title={Transparent Adaptation in Deep Medical Image Diagnosis.}, author={Kollias, Dimitris and Vlaxos, Y and Seferis, M and Kollia, Ilianna and Sukissian, Levon and Wingate, James and Kollias, Stefanos D}, booktitle={TAILOR}, pages={251–267}, year={2020}}

• D. Kollias, et. al.: “Deep neural architectures for prediction in healthcare”, CIS, 2018 

@article{kollias2018deep, title={Deep neural architectures for prediction in healthcare}, author={Kollias, Dimitrios and Tagaris, Athanasios and Stafylopatis, Andreas and Kollias, Stefanos and Tagaris, Georgios}, journal={Complex \& Intelligent Systems}, volume={4}, number={2}, pages={119–131}, year={2018}, publisher={Springer}}