Automated erythrocyte detection and classification from whole slide images

Blood smear is a crucial diagnostic aid. Quantification of both solitary and overlapping erythrocytes within these smears, directly from their whole slide images (WSIs), remains a challenge.  Existing software designed to accomplish the computationally extensive task of hematological whole slide image analysis are too expensive and are widely unavailable. We have thereby developed a fully automated software targeted for erythrocyte detection and quantification from WSIs. In this paper, we define an optimal region within the smear which contain cells which are neither too scarce or damaged nor too crowded. We detect the optimal regions within the smear and subsequently extract all the cells from these regions, both solitary and overlapped, the latter of which undergoes a clump splitting (see the figure below) before extraction. The performance was systematically tested on 28 WSIs of blood smears obtained from 13 different species from three classes of the subphylum vertebrata including birds, mammals, and reptiles. This data poses as an immensely variant erythrocyte database with diversity in size, shape, intensity and textural features. The results suggest the potential employment of this software for the diagnosis of hematological disorders like sickle cell anemia.

(Select result) Clump splitting pipeline
(Select result) Clump splitting pipeline. (A) Clumped cells in gray scale with local image patch highlighted. (B) Line drawn connecting the bottleneck points in the image patch to obtain the angle of tilt. (C) Vertical alignment of the image patch. (D) Horizontal line profile obtained along the mid-section of the image patch is shown in red. Corresponding maximum likelihood fit to determine the locations of local intensity peaks is shown in blue. (E) Points depicting the boundaries of the overlapped region: bottleneck (R* and Q*) and side points (S and T). (F) Binary image of the overlapped cells with convex hull highlighted. (G) Subtraction of convex hull from original image. (H) Segmented cells.

Selected References

D. Govind, B. Lutnick, J. E. Tomaszewski, and P. Sarder, “Automated erythrocyte detection and classification from whole slide images,” Journal of Medical Imaging – SPIE, vol. 5, no. 2, pp. 027501:1-11, Apr. 2018.


Underline indicates corresponding author.
* indicates equal contribution.
indicates Dr. Sarder’s faculty trainees.
§ indicates post doctoral associates.
indicates Dr. Sarder’s graduate students.
indicates Dr. Sarder’s undergraduate students.