A De-Identification Pipeline for Ultrasound Medical Images in DICOM Format
Clinical data sharing between healthcare institutions, and between practitioners is often hindered by privacy protection requirements. This problem is critical in collaborative scenarios where data sharing is fundamental for establishing a workflow among parties. The anonymization of patient information burned in DICOM images requires elaborate processes somewhat more complex than simple de-identification of textual information. Usually, before sharing, there is a need for manual removal of specific areas containing sensitive information in the images. In this paper, we present a pipeline for ultrasound medical image de-identification, provided as a free anonymization REST service for medical image applications, and a Software-as-a-Service to streamline automatic de-identification of medical images, which is freely available for end-users. The proposed approach applies image processing functions and machine-learning models to bring about an automatic system to anonymize medical images. To perform character recognition, we evaluated several machine-learning models, being Convolutional Neural Networks (CNN) selected as the best approach. For accessing the system quality, 500 processed images were manually inspected showing an anonymization rate of 89.2%. The tool can be accessed at https://bioinformatics.ua.pt/dicom/anonymizer and it is available with the most recent version of Google Chrome, Mozilla Firefox and Safari. A Docker image containing the proposed service is also publicly available for the community.
KeywordsMedical imaging OCR Neural networks Deep-learning De-identification
Compliance with Ethical Standards
This work was supported by project Cloud Thinking (CENTRO-07-ST24-FEDER-002031), co-funded by QREN, “Mais Centro” program, and the EU/EFPIA Innovative Medicines Initiative Joint Undertaking (EMIF grant n° 115,372). Eriksson Monteiro is funded by Fundação para a Ciência e Tecnologia (FCT) under the grant agreement SFRH/BD/102195/2014.
Conflict of Interest
All authors declare that there are no conflicts of interest in this work.
This article does not contain any studies with human participants or animals performed by any of the authors.
- 2.Somolinos, R., Munoz, A., Hernando, M.E., Pascual, M., Caceres, J., Sanchez-de-Madariaga, R., Fragua, J.A., Serrano, P., and Salvador, C.H., Service for the Pseudonymization of electronic healthcare records based on ISO/EN 13606 for the secondary use of information. IEEE J. Biomed. Heal. Informatics. 19(6):2015, 1937–1944.Google Scholar
- 4.Clunie, D., How to use DoseUtilityTM. PixelMed Publishing. Available: http://www.dclunie.com/pixelmed/software/webstart/DoseUtilityUsage.html. Accessed 26 Jul 2016.
- 9.Huang, H. K., PACS and imaging informations: Basic principles and applications. Wiley-Blackwell, 2004.Google Scholar
- 12.Shahbaz, S., Mahmood, A., and Anwar, Z., SOAD: Securing oncology EMR by anonymizing DICOM images. In: Proceedings -11th International Conference on Frontiers of Information Technology, FIT 2013, 2013, pp. 125–130.Google Scholar
- 14.Rodríguez González, D., Carpenter, T., van Hemert, J.I., and Wardlaw, J., An open source toolkit for medical imaging de-identification. Eur. Radiol. 20(8):2010, 1896–1904.Google Scholar
- 16.Li, L. and Wang, J. Z., DDIT - A Tool for DICOM Brain Images De-Identification. In: 2011 5th International Conference on Bioinformatics and Biomedical Engineering, 2011, pp. 1–4.Google Scholar
- 19.Florea, F., Rogozan, A., and Bensrhair, A., Modality categorization by textual annotations interpretation in medical imaging. Med. Informatics Eur. (MIE 2005) :1270–1275, 2005.Google Scholar
- 20.Chambolle, A., An algorithm for Total variation minimization and applications. Journal of Mathematical Imaging and Vision. 20(1–2):89–97, 2004.Google Scholar
- 21.Bradski, G. and Kaehler, A., Learning OpenCV: Computer Vision with the OpenCV Library. Vol 1. 2008.Google Scholar
- 23.Community, O., The OpenCV reference manual. October. 1–1104, 2010.Google Scholar
- 25.de Campos, T. E., Babu, B. R., and Varma, M., Character recognition in natural images. Proc. Int. Conf. Comput. Vis. Theory Appl. 2009.Google Scholar
- 27.T. Tieleman, Training Restricted Boltzmann Machines using Approximations to the Likelihood Gradient. Proc. 25th Int. Conf. Mach. Learn. 307: 7, 2008.Google Scholar
- 28.Larochelle, H., Mandel, M., Pascanu, R., and Bengio, Y., Learning algorithms for the classification Restricted Boltzmann machine. J. Mach. Learn. Res. 13:643–669, 2012.Google Scholar
- 30.Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, É., Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12:2825–2830, 2012.Google Scholar
- 31.Krizhevsky A., Sutskever I., and Hinton, G. E., ImageNet Classification with Deep Convolutional Neural Networks. In Advances in Neural Information Processing Systems, 2012, pp. 1097–1105.Google Scholar
- 34.Bergstra, J., Bastien, F., Breuleux, O., Lamblin, P., Pascanu, R., Delalleau, O., Desjardins, G., Warde-Farley, D., Goodfellow, I., Bergeron, A., and Bengio, Y., Theano: Deep learning on GPUs with Python. J. Mach. Learn. Res. 1:1–48, 2011.Google Scholar
- 35.Levenshtein, V.I., Binary codes capable of correcting deletions. Insertions and Reversals. Sov. Phys. Dokl. 10:707, 1966.Google Scholar
- 36.BMD software, PACScenter. Available: https://demo.bmd-software.com/viewer. Accessed 26 Jul 2016.
- 37.Melicio Monteiro, E. J., Costa, C., and Oliveira, J. L., A DICOM viewer based on web technology. In: 2013 I.E. 15th International Conference on e-Health Networking, Applications and Services (Healthcom 2013), 2013, pp. 167–171.Google Scholar