An End-to-end System for Automatic Characterization of Iba1 Immunopositive Microglia in Whole Slide Imaging
Traumatic brain injury (TBI) is one of the leading causes of death and disability worldwide. Detailed studies of the microglial response after TBI require high throughput quantification of changes in microglial count and morphology in histological sections throughout the brain. In this paper, we present a fully automated end-to-end system that is capable of assessing microglial activation in white matter regions on whole slide images of Iba1 stained sections. Our approach involves the division of the full brain slides into smaller image patches that are subsequently automatically classified into white and grey matter sections. On the patches classified as white matter, we jointly apply functional minimization methods and deep learning classification to identify Iba1-immunopositive microglia. Detected cells are then automatically traced to preserve their complex branching structure after which fractal analysis is applied to determine the activation states of the cells. The resulting system detects white matter regions with 84% accuracy, detects microglia with a performance level of 0.70 (F1 score, the harmonic mean of precision and sensitivity) and performs binary microglia morphology classification with a 70% accuracy. This automated pipeline performs these analyses at a 20-fold increase in speed when compared to a human pathologist. Moreover, we have demonstrated robustness to variations in stain intensity common for Iba1 immunostaining. A preliminary analysis was conducted that indicated that this pipeline can identify differences in microglia response due to TBI. An automated solution to microglia cell analysis can greatly increase standardized analysis of brain slides, allowing pathologists and neuroscientists to focus on characterizing the associated underlying diseases and injuries.
KeywordsComputer-aided detection and diagnosis Whole slide imaging Digital pathology Image analysis Classification Traumatic brain injury
This work was supported in part by funding from Natural Sciences and Engineering Research Council of Canada (NSERC), Canadian Institutes of Health Research (CIHR), the BC Innovation Council NRAS Program, and the Weston Brain Institute. Support from Professor Salcudean’s C. A. Laszlo Chair is gratefully acknowledged. Dr. Nir is a recipient of a Prostate Cancer Canada Post-Doctoral Research Fellowship Award #PDF2016-1338.
Compliance with Ethical Standards
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance with the ethical standards of the institution or practice at which the studies were conducted.
Conflict of interests
The authors declare that they have no conflict of interest.
Information Sharing Statement
The data and source code utilized in this work are available at https://www.frdr.ca/repo/handle/doi:10.20383/101.0120/ and https://github.com/rcldigitalpathology/Brain-Analyzer/, respectively.
- Abdolhoseini, M., Walker, F., Johnson, S. (2016). Automated tracing of microglia using multilevel thresholding and minimum spanning trees. In 2016 IEEE 38th annual international conference of the engineering in medicine and biology society (EMBC) (pp. 1208–1211). IEEE.Google Scholar
- Balis, U., Hipp, J., Flotte, T., Monaco, J., Cheng, J., Madabhushi, A., Yagi, Y., Rodriguez-Canales, J., Emmert-Buck, M., Dugan, M., et al. (2011). Computer aided diagnostic tools aim to empower rather than replace pathologists: lessons learned from computational chess. Journal of Pathology Informatics, 2(1), 25. https://doi.org/10.4103/2153-3539.82050.CrossRefGoogle Scholar
- Bishop, C.M. (2006). Pattern recognition and machine learning (information science and statistics), 1st edn. 2006. corr. 2nd printing edn. New York: Springer.Google Scholar
- Cheng, W.H., Stukas, S., Martens, K.M., Namjoshi, D.R., Button, E.B., Wilkinson, A., Bashir, A., Robert, J., Cripton, P.A., Wellington, C.L. (2018). Age at injury and genotype modify acute inflammatory and neurofilament-light responses to mild chimera traumatic brain injury in wild-type and app/ps1 mice. Experimental Neurology, 301, 26–38.CrossRefGoogle Scholar
- Cherry, J.D., Tripodis, Y., Alvarez, V.E., Huber, B., Kiernan, P.T., Daneshvar, D.H., Mez, J., Montenigro, P.H., Solomon, T.M., Alosco, M.L., et al. (2016). Microglial neuroinflammation contributes to tau accumulation in chronic traumatic encephalopathy. Acta Neuropathologica Communications, 4(1), 112.CrossRefGoogle Scholar
- Cireşan, D., Giusti, A., Gambardella, L.M., Schmidhuber, J. (2012). Deep neural networks segment neuronal membranes in electron microscopy images. In Advances in neural information processing systems (pp. 2843–2851).Google Scholar
- Costa, A. (2013). Hausdorff (box-counting) fractal dimension. https://www.mathworks.com/matlabcentral/fileexchange/30329-hausdorff-box-counting-fractal-dimension/, [Online; accessed March 2017].
- Coughlin, J.M., Wang, Y., Minn, I., Bienko, N., Ambinder, E.B., Xu, X., Peters, M.E., Dougherty, J.W., Vranesic, M., Koo, S.M., et al. (2017). Imaging of glial cell activation and white matter integrity in brains of active and recently retired national football league players. JAMA Neurology, 74(1), 67–74.CrossRefGoogle Scholar
- Cruz-Roa, A., Basavanhally, A., González, F, Gilmore, H., Feldman, M., Ganesan, S., Shih, N., Tomaszewski, J., Madabhushi, A. (2014). Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks. In Medical imaging 2014: digital pathology, international society for optics and photonics (Vol. 9041, p. 904103).Google Scholar
- Duda, R.O., Hart, P.E., Stork, D.G. (2012). Pattern classification. New York: Wiley.Google Scholar
- Ertosun, M.G., & Rubin, D.L. (2015). Automated grading of gliomas using deep learning in digital pathology images: a modular approach with ensemble of convolutional neural networks. In AMIA annual symposium proceedings, American medical informatics association (Vol. 2015, pp. 1899–1908).Google Scholar
- Farahani, N., Parwani, A.V., Pantanowitz, L. (2015). Whole slide imaging in pathology: advantages, limitations, and emerging perspectives. Journal of Pathology and Laboratory Medicine International, 7, 23–33.Google Scholar
- Freund, Y., Schapire, R., Abe, N. (1999). A short introduction to boosting. Journal of Japanese Society for Artificial Intelligence, 14(5), 771–780.Google Scholar
- Gonzalez, R.C., & Woods, R.E. (2008). Digital image processing. Upper Saddle River: Pearson-Prentice-Hall.Google Scholar
- Kaplan, K.J., & Rao, L.K. (2015). Digital pathology: historical perspectives, current concepts & future applications. Cham: Springer.Google Scholar
- Kollmannsberger, P. (2009). Skeleton3d. https://www.mathworks.com/matlabcentral/fileexchange/43400-skeleton3d/, [Online; accessed March 2017].
- Krizhevsky, A., Sutskever, I., Hinton, G.E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems (pp. 1097–1105).Google Scholar
- Madabhushi, A., & Lee, G. (2016). Image analysis and machine learning in digital pathology: challenges and opportunities.Google Scholar
- McKee, A.C., Cantu, R.C., Nowinski, C.J., Hedley-Whyte, E.T., Gavett, B.E., Budson, A.E., Santini, V.E., Lee, H.S., Kubilus, C.A., Stern, R.A. (2009). Chronic traumatic encephalopathy in athletes: progressive tauopathy after repetitive head injury. Journal of Neuropathology & Experimental Neurology, 68(7), 709–735.CrossRefGoogle Scholar
- Namjoshi, D.R., Cheng, W.H., McInnes, K.A., Martens, K.M., Carr, M., Wilkinson, A., Fan, J., Robert, J., Hayat, A., Cripton, P.A., et al. (2014). Merging pathology with biomechanics using chimera (closed-head impact model of engineered rotational acceleration): a novel, surgery-free model of traumatic brain injury. Molecular Neurodegeneration, 9(1), 55.CrossRefGoogle Scholar
- Namjoshi, D.R., Cheng, W.H., Bashir, A., Wilkinson, A., Stukas, S., Martens, K.M., Whyte, T., Abebe, Z.A., McInnes, K.A., Cripton, P.A., et al. (2017). Defining the biomechanical and biological threshold of murine mild traumatic brain injury using chimera (closed head impact model of engineered rotational acceleration). Experimental Neurology, 292, 80–91.CrossRefGoogle Scholar
- Nir, G., Hor, S., Karimi, D., Fazli, L., Skinnider, B.F., Tavassoli, P., Turbin, D., Villamil, C.F., Wang, G., Wilson, R.S., et al. (2018). Automatic grading of prostate cancer in digitized histopathology images: learning from multiple experts. Medical Image Analysis, 50(12), 167–180.CrossRefGoogle Scholar
- Ramlackhansingh, A.F., Brooks, D.J., Greenwood, R.J., Bose, S.K., Turkheimer, F.E., Kinnunen, K.M., Gentleman, S., Heckemann, R.A., Gunanayagam, K., Gelosa, G., et al. (2011). Inflammation after trauma: microglial activation and traumatic brain injury. Annals of Neurology, 70(3), 374–383.CrossRefGoogle Scholar
- Rashid, S., Nir, G., Fazli, L., Boag, A.H., Siemens, D.R., Goldenberg, S.L., Abolmaesumi, P., Salcudean, S.E. (2018). Automatic pathology of prostate cancer in whole mount slides incorporating individual gland classification. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization. https://doi.org/10.1080/21681163.2018.1514280.
- Steinwart, I., & Christmann, A. (2008). Support vector machines. Berlin: Springer Science & Business Media.Google Scholar
- Strekalovskiy, E., & Cremers, D. (2014). Real-time minimization of the piecewise smooth mumford-shah functional. In European conference on computer vision (pp. 127–141). Springer.Google Scholar
- The MathWorks. (2017a). Matlab: Deep learning. https://www.mathworks.com/help/nnet/examples/create-simple-deep-learning-network-for-classification.html/ [Online; accessed 1 December 2017].
- The MathWorks. (2017b). Matlab: Regional maxima. https://www.mathworks.com/help/images/ref/imregionalmax.html/ [Online; accessed 1 December 2017].
- Vadakkan, T.J. (2009). Lacunarity of a binary image. https://www.mathworks.com/matlabcentral/fileexchange/25261-lacunarity-of-a-binary-image/, [Online; accessed August 2018].
- Van Gool, L., Szekely, G., Ferrari, V. (2014). Computer Vision.Google Scholar