10.14489/td.2026.10.pp.076-082

2026, 10 October

DOI: 10.14489/td.2026.10.pp.076-082

Egorchev A. A., Rosin A. A., Asaulenko Z. P., Chikrin D. E., Paveliev M. N.
CASCADED ALGORITHMIC SUPPORT FOR AN INFORMATION-MEASUREMENT SYSTEM OF A SCANNING MICROSCOPE FOR PROCESSING WHOLE-SLIDE BONE-MARROW IMAGES AND ANALYSING MEGAKARYOCYTES
(pp. 76-82)

Abstract. This paper proposes cascaded algorithmic support for an information and measurement system of a slide-scanning microscope intended for automated analysis of megakaryocytes in bone marrow whole-slide images. Cascaded processing reduces the computational burden associated with gigapixel images and produces an object-level representation suitable for subsequent morphometric analysis. At a lower level of the image pyramid, a neural network generates a megakaryocyte probability map and identifies anchor points of candidate objects. For each point, a 384×384-pixel region is extracted at the highest-resolution level, where cell segmentation and validation are performed. The resulting mask, bounding box, centroid, and calculated features are transformed into the global coordinate system of the whole-slide image and stored in an object table. The dataset comprises 40 whole-slide images, 37.410 point annotations, 3,983 cell polygons, and 4.844 nucleus polygons. For the selected test image, point localization recall was 0.810, point coverage by local regions was 0.917, and coverage of polygon-annotated cells was 0.939. In an independent evaluation of the local stage using 99 reference regions, the mean Dice coefficient and intersection over union were 0.852 and 0.759, respectively. The results confirm the applicability of the proposed algorithmic support to object-level representation and quantitative analysis of megakaryocytes.

Keywords: information and measurement system, slide-scanning microscopy, whole-slide images, bone marrow, megakaryocytes, cascaded image processing, instance segmen-tation, object-level representation, morphometric analysis, neural networks.

+ - About the Authors Click to collapse

A. A. Egorchev, A. A. Rosin, Z. P. Asaulenko, D. E. Chikrin, M. N. Paveliev (Institute of Computational Mathematics and Information Technology, Kazan Federal University, Kazan, Russia) E-mail: Данный адрес e-mail защищен от спам-ботов, Вам необходимо включить Javascript для его просмотра. , Данный адрес e-mail защищен от спам-ботов, Вам необходимо включить Javascript для его просмотра. , Данный адрес e-mail защищен от спам-ботов, Вам необходимо включить Javascript для его просмотра. , Данный адрес e-mail защищен от спам-ботов, Вам необходимо включить Javascript для его просмотра. , Данный адрес e-mail защищен от спам-ботов, Вам необходимо включить Javascript для его просмотра.

+ - References Click to collapse

1. Sheremeteva, T. A., Malov, A. M., & Filippov, G. N. (2009). Image processing and morphometric measurements of objects in microscopy. Izvestiya vysshikh uchebnykh zavedeniy. Priborostroenie, 52(8), 68–73. [in Russian language].
2. Xu, H., Usuyama, N., Bagga, J., et al. (2024). A whole-slide foundation model for digital pathology from real-world data. Nature, 630(8015), 181–188.
3. Chen, R. J., Ding, T., Lu, M. Y., et al. (2024). Towards a general-purpose foundation model for computational pathology. Nature Medicine, 30, 850–862.
4. Ghete, T., Kock, F., Pontones, M., et al. (2024). Models for the marrow: A comprehensive review of AI-based cell classification methods and malignancy detection in bone marrow aspirate smears. HemaSphere, 8, Article e70048.
5. Wang, X., Wang, Y., Qi, C., et al. (2023). The application of Morphogo in the detection of megakaryocytes from bone marrow digital images with convolutional neural networks. Technology in Cancer Research & Treatment, 22(7).
6. Tayebi, R. M., Mu, Y., Dehkharghanian, T., et al. (2022). Automated bone marrow cytology using deep learning to generate a histogram of cell types. Communications Medicine, 2(1).
7. Qu, H., Wu, P., Huang, Q., et al. (2020). Weakly supervised deep nuclei segmentation using partial points annotation in histopathology images. IEEE Transactions on Medical Imaging, 39(11), 3655–3666.
8. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 (Vol. 9135, pp. 234–241). Springer.
9. He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask R-CNN. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 2961–2969). IEEE.
10. Jha, A., Yang, H., Deng, R., et al. (2021). Instance segmentation for whole slide imaging: End-to-end or detect-then-segment. Journal of Medical Imaging, 8(1), Article 014001.
11. Ma, J., He, Y., Li, F., et al. (2024). Segment anything in medical images. Nature Communications, 15, Article 654.
12. Stringer, C., Wang, T., Michaelos, M., & Pachitariu, M. (2021). Cellpose: A generalist algorithm for cellular segmentation. Nature Methods, 18, 100–106.
13. Tellez, D., Litjens, G., Bándi, P., et al. (2019). Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology. Medical Image Analysis, 58, Article 101544.

+ - Purchase digital version of a single article Click to collapse

This article is available in electronic format (PDF).

DOI: 10.14489/td.2026.10.pp.076-082

Copy the article DOI and follow the link https://id-spektr.ru/product/pokupka-elektronnoy-stati-iz-zhurnala-kontrol-diagnostika

Please specify the article DOI in the order comments.