HEp-2 Immuno-Fluorescence (IIF) Images Classification and Segmentation with Random Forests


Abstract:


Human epithelial (HEp-2) cell specimens is obtained from indirect immunofluorescence (IIF) imaging for diagnosis and management of autoimmune diseases. Analysis of HEp2 cells is important and in this work we consider automatic cell segmentation and classification using spatial and texture pattern features and random forest (RF) classifiers. In this paper, we summarize our efforts in classification and segmentation tasks proposed in ICPR 2016 contest.

Task 1: Cell level staining pattern classification

For the cell level staining pattern classification (Task 1), we utilized texture features such as rotational invariant co-occurrence (RIC) versions of the well-known local binary pattern (LBP), median binary pattern (MBP), joint adaptive median binary pattern (JAMBP), and motif labels (ML) along with other optimized features. We report the classification results utilizing different classifiers such as the k-nearest neighbors (kNN), support vector machine (SVM), and random forest (RF). We obtained the best accuracy of 94.26% for six cell classes with RIC-LBP combined with a motif co-occurrence labels (MCL).


Task 2: Specimen level staining pattern classification


For specimen level staining pattern classification (Task 2) we utilize a combination RIC-LBP with RF classifier and obtained 80% accuracy for seven classes.


Task 4: Cell segmentation


For cell segmentation (Task 4), we use our optimized multiscale spatial feature bank along with RF classifier for pixelwise labeling and obtained F-measure of 84.26% for 1008 images.


Example segmentations from our pipeline

(Top to bottom: Input, ground-truth (GT), our results)

For executable code requests please email the corresponding authors given below:

Task 1: Cell level staining pattern classificaiton (Mr. Zakariya Oraibi - zaonr5@mail.missouri.edu)

Task 2: Specimen level staining pattern classification (Mr. Zakariya Oraibi - zaonr5@mail.missouri.edu and Mr. Jean-Baptiste Guiriec - jeanbaptiste.guiriec@insa-cvl.fr)

Task 4: Cell segmentation - (Ms. Yasmin Kassim - ymkgz8@mail.missouri.edu)


Reference:

V. B. S. Prasath, Y. M. Kassim, Z. A. Oraibi, J.-B. Guiriec, A. Hafiane, K. Palaniappan. HEp-2 cell classification and segmentation using motif texture patterns and spatial features with random forests. 23rd International Conference on Pattern Recognition, International Contest on Pattern Recognition Techniques for Indirect Immunofluorescence Images Analysis, Cancun, Mexico. Proc. IEEE, pp. 90-95, December 2016. doi:10.1109/ICPR.2016.7899614

For a more comprehensive work on HEp-2 cell segmentation, please see HEp-2SegZoo:

B. Iyer, S. Deoghare, K. Ranjan, B. J. Aronow, V. B. S. Prasath. Benchmarking HEp-2 cell segmentation methods in indirect immunofluorescence images - Standard models to deep learning. Submitted, 2024. [Code]

Back Microscopy Projects Main page. Back to Research Page.