Asad Ullah


AbstractIndirect Immuno Fluorescence (IIF) detection analysis technique is in limelight because of its great importance in the field of medical health. It is mainly used for the analysis of auto-immune diseases. These diseases are caused when body’s natural defense system can’t distinguish between normal body cells and foreign cells. More than 80 auto-immune diseases exist in humans which affect different parts of body. IIF works both manually as well as by using Computer-Aided Diagnosis (CAD). The aim of research is to propose an advanced methodology for the analysis of auto-immune diseases by using well-known model of transfer learning for the analysis of autoimmune diseases. Data augmentation and data normalization is also used to resolve the problem of over fitting in data. Firstly, freely available MIVIA data set of HEP- type 2 cells has been selected, which contains total of 1457 images and six different classes of staining patterns named as centromere, homogeneous, nucleolar, coarse speckled, fine speckled and cytoplasmatic. Then well-known model of transfer learning VGG-16 are train on MIVIA data set of HEP-type 2 cells. Data augmentation and data normalization used on pre-trained data to avoid over fitting because datasets of medical images are not very large. After the application of data augmentation and data normalization on pre-trained model, the performance of model is used to calculate by using a confusion matrix of VGG-16. VGG-16 achieves 84.375% accuracy. It is more suitable for the analysis of auto-immune diseases. Same as accuracy, we also use the other three parameters, Precision, F1 measures, and recall to check the performance of model. All four parameters use confusion matrix to find performance of model. The tools and languages also have great importance because it gives a simple and easy way of implementation to solve problems in image processing. For this purpose, python and colab is used to read and write the data because python provides fast execution of data and colab work as a simulator of python. The result shows that transfer learning is the most sufficient and enhanced technique for the analysis of auto-immune diseases since it provides high accuracy in less time and reduces the errors in image classification.

Full Text:



Grygiel-Górniak, B., Rogacka, N., &Puszczewicz, M. (2018). Antinuclear antibodies in healthy people and non- rheumatic diseases–diagnostic and clinical implications. Reumatologia, 56(4), 243.

Agmon-Levin, N., Damoiseaux, J., Kallenberg, C., Sack, U., Witte, T., Herold, M., ... &Shoenfeld, Y. (2014). International recommendations for the assessment of autoantibodies to cellular antigens referred to as anti-nuclear antibodies. Annals of the rheumatic diseases, 73(1), 17-23.

Bizzaro, N., Antico, A., Platzgummer, S., Tonutti, E., Bassetti, D., Pesente, F., ... & Study Group on Autoimmune Diseases of the Italian Society of Laboratory Medicine. (2014). Automated antinuclear immunofluorescence antibody screening: a comparative study of six computer-aided diagnostic systems. Autoimmunity Reviews, 13(3), 292-298.

Hiemann, R., Büttner, T., Krieger, T., Roggenbuck, D., Sack, U., & Conrad, K. (2009). Challenges of automated screening and differentiation of non-organ specific autoantibodies on HEp-2 cells. Autoimmunity reviews, 9(1), 17- 22.

Rigon, A., Buzzulini, F., Soda, P., Onofri, L., Arcarese, L., Iannello, G., &Afeltra, A. (2011). Novel opportunities in automated classification of antinuclear antibodies on HEp-2 cells. Autoimmunity Reviews, 10(10), 647-652.

Willitzki, A., Hiemann, R., Peters, V., Sack, U., Schierack, P., Rödiger, S., ... & Roggenbuck, D. (2012). New platform technology for comprehensive serological diagnostics of autoimmune diseases. Clinical and Developmental Immunology, 2012.

Vivona, L., Cascio, D., Taormina, V., &Raso, G. (2018). Automated approach for indirect immunofluorescenceimages classification based on unsupervised clustering method. IET Computer Vision, 12(7), 989-995.

Donato, C., Vincenzo, T., Marco, C., Francesco, F., Maria, V. S., & Giuseppe, R. (2014, August). HEp-2 cell classification with heterogeneous classes-processes based on k-nearest neighbours. In 2014 1st Workshop onPattern Recognition Techniques for Indirect Immunofluorescence Images (pp. 10-15). IEEE.

BenammarElgaaied, A., Cascio, D., Bruno, S., Ciaccio, M. C., Cipolla, M., Fauci, A., ... &Raso, G. (2016). Computer-assisted classification patterns in autoimmune diagnostics: the AIDA project. BioMed research international, 2016.

Ciatto, S., Cascio, D., Fauci, F., Magro, R., Raso, G., Ienzi, R., ... & Simone, M. V. (2009). Computer-assiste diagnosis (CAD) in mammography: comparison of diagnostic accuracy of a new algorithm (Cyclopus®,Medicad) with two commercial systems. La radiologia medica, 114(4), 626-635.

Cascio, D., Fauci, F., Iacomi, M., Raso, G., Magro, R., Castrogiovanni, D., ... &Vasile, M. S. (2014). Computer-aided diagnosis in digital mammography: comparison of two commercial systems. Imaging in Medicine, 6(1), 13-20.

Foggia, P., Percannella, G., Soda, P., & Vento, M. (2013). Benchmarking HEp-2 cells classification methods. IEEE transactions on medical imaging, 32(10), 1878-1889.

Foggia, P., Percannella, G., Saggese, A., & Vento, M. (2014). Pattern recognition in stained hep-2 cells: Whereare we now? Pattern Recognition, 47(7), 2305-2314.

Cascio, D., Taormina, V., &Raso, G. (2019). Deep CNN for IIF images classification in autoimmune diagnostics. Applied Sciences, 9(8), 1618.


  • There are currently no refbacks.

ISSN 2307-552X