Rhino-Cyt: A System for Supporting the Rhinologist in the Analysis of Nasal Cytology
In recent years, cytological observations in the rhinological field are being increasingly utilized. This development has taken place over the last two decades and has proven to be fundamental in defining new nosological entities and in driving changes in the previous classification of rhinopathies. The simplicity of the technique and its low invasiveness make rhinocytology a practical diagnostic tool practical for all rhinoallergology services. Furthermore, since it allows the monitoring of responses to treatment, this method plays an important role in guiding a more effective and less expensive diagnostic program. Microscopic observation requires prolonged effort by a specialist, but the modern scanning systems for cytological preparations allow scanning of an entire preparation enlarged to 400x. By means of the system presented in this paper, it is possible to automatically identify and classify cells present on a rhinocytologic preparation based on a digital image of the preparation itself. Thus, pivotal diagnostic support has been made available to the rhinocytologist, who can quickly verify that the cells have been correctly classified by observation on a monitor. In the system presented herein, image processing and image segmentation techniques have been used to find images of cellular elements within the preparation. Cell classification is based on a convolutional neural network composed of three blocks of main layers. Cell identification (first step, image segmentation) exhibits accuracy greater than 90%, while cell classification (second step, seven cytotypes) attained a mean accuracy of approximately 98%. Finally, the classified cell images are shown to a specialist for rapid verification. This complete system supports clinicians in the preparation of a rhinocytogram report.
KeywordsNasal cytology Automatic cell recognition Rhinologic Image analysis
We thank the Dr. Alfredo Zito, head of the Department of Pathological Anatomy of I.R.C.C.S., for making the D-Sight available to scan the preparations used.
- 1.Gelardi, M.: Atlas of Nasal Cytology for the Differential Diagnosis of Nasal Diseases. Edi. Ermes, Milano (2012)Google Scholar
- 2.Piuri, V., Scotti, F.: Morphological classification of blood leucocytes by microscope images. In: 2004 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications, CIMSA 2004, pp. 103–108 (2004)Google Scholar
- 5.Bevilacqua, V., Buongiorno, D., Carlucci, P., Giglio, F., Tattoli, G., Guarini, A., Sgherza, N., De Tullio, G., Minoia, C., Scattone, A., Simone, G., Girardi, F., Zito, A., Gesualdo, L.: A supervised CAD to support telemedicine in hematology. In: Proceedings of the International Joint Conference on Neural Networks (2015)Google Scholar
- 6.Python 3.6.5: https://docs.python.org/3. Accessed 03 May 2018
- 7.Pycharm: https://www.jetbrains.com/pycharm/documentation/. Accessed 03 May 2018
- 8.Keras: https://keras.io/. Accessed 03 May 2018
- 9.Hyperas: https://github.com/maxpumperla/hyperas. Accessed 03 May 2018
- 10.Scipy: https://www.scipy.org/docs.html. Accessed 03 May 2018
- 12.Agarap, A.F.: An architecture combining convolutional neural network (CNN) and support vector machine (SVM) for image classification. arXiv1712.03541 (2017)Google Scholar