A Two-Stage Automated Framework for Quantitative Morphological Analysis of Mast Cell Degranulation in Histopathological Images Based on YOLO and CNN

Enna Chen; Jiadong Li; Xuan Qiao; Liujie Ren; Zouqin Huang; Wei Yao; Yi Yu

doi:10.5566/ias.3751

Authors

Enna Chen Shanghai Key Laboratory of Acupuncture Mechanism and Acupoint Function, College of Biomedical Engineering, Fudan University
Jiadong Li Shanghai Key Laboratory of Acupuncture Mechanism and Acupoint Function, School of Biomedical Engineering and Technology Innovation, Fudan University
Xuan Qiao Shanghai Key Laboratory of Acupuncture Mechanism and Acupoint Function, School of Biomedical Engineering and Technology Innovation, Fudan University
Liujie Ren Eye and ENT Hospital of Fudan University
Zouqin Huang Shanghai Pudong New Area Traditional Chinese Medicine Hospital
Wei Yao Shanghai Key Laboratory of Acupuncture Mechanism and Acupoint Function, School of Biomedical Engineering and Technology Innovation, Fudan University
Yi Yu College of Medical Instruments, Shanghai University of Medicine & Health Sciences

DOI:

https://doi.org/10.5566/ias.3751

Keywords:

Class Activation Mapping, CNN, Degranulation, Mast Cell, Quantitative Image Analysis, YOLO

Abstract

Quantitative analysis of mast cell (MC) morphology and degranulation states is crucial for assessing inflammatory responses and therapeutic efficacy in biomedical research. This study presents a novel two-stage deep learning framework for the automated quantitative morphological analysis of MC degranulation states in toluidine blue-stained histological sections. We constructed a specialized dataset of 1,054 rat tissue images. In the detection stage, YOLOv11m achieved superior performance with a mean average precision (mAP@0.5) of 84% for locating MCs amidst complex tissue backgrounds. In the classification stage, using the model we previously acquired to extract pure mast cell images, EfficientNetV2-S attained an accuracy of 89.6% ± 2.1% in discriminating degranulation states through fine-grained morphological analysis. Critically, Class Activation Mapping (CAM) visualization demonstrated that the model’s decision logic aligns precisely with pathological features of degranulation—such as membrane rupture and granule dispersal—thereby providing interpretable morphological evidence for automated classification. The proposed framework effectively decouples the tasks of cell localization and state classification, offering a robust, efficient, and morphologically interpretable solution for quantitative image analysis in histopathology. This approach has significant applications in acupuncture mechanism research and can be extended to other fields requiring granular structure analysis.

References

Abd El-Aziz AA, Mahmood MA, Abd El-Ghany S (2025). A robust EfficientNetV2-S classifier for predicting acute lymphoblastic leukemia based on cross validation. Symmetry Basel 17:24.

Albert-Bayo M, Paracuellos I, Gonzalez-Castro AM, Rodriguez-Urrutia A, Rodriguez-Lagunas MJ, Alonso-Cotoner C, Santos J, Vicario M (2019). Intestinal mucosal mast cells: Key modulators of barrier function and homeostasis. Cells 8:135.

Anaam A, Al-antari MA, Hussain J, Samee NA, Alabdulhafith M, Gofuku A (2023). Deep active learning for automatic mitotic cell detection on HEp-2 specimen medical images. Diagnostics 13:1416.

Austen KF, Boyce JA (2001). Mast cell lineage development and phenotypic regulation. Leuk Res 25:511–8.

Cao JB, Zhu ST, Huang XS, Wang XY, Wu ML, Li X, Liu FL, Chen L, Zheng YT, Wang JH (2024). Mast cell degranulation-triggered by SARS-CoV-2 induces tracheal-bronchial epithelial inflammation and injury. Virol Sin 39:309–18.

Du X, Liu L, Wang X, Ni G, Zhang J, Hao R, Liu J, Liu Y (2019). Automatic classification of cells in microscopic fecal images using convolutional neural networks. Biosci Rep 39:BSR20182100.

Gaudenzio N, Sibilano R, Marichal T, Starkl P, Reber LL, Cenac N, McNeil BD, Dong X, Hernandez JD, Sagi-Eisenberg R, Hammel I, Roers A, Valitutti S, Tsai M, Espinosa E, Galli SJ (2016). Different activation signals induce distinct mast cell degranulation strategies. J Clin Invest 126:3981–98.

Gual-Vaya L (2024). Classification of red blood cells from a geometric morphometric study. Image Anal Stereol 43:109–19.

Guo X, Sun M, Yang P, Meng X, Liu R (2023). Role of mast cells activation in the tumor immune microenvironment and immunotherapy of cancers. Eur J Pharmacol 960:176103.

Haq I, Mazhar T, Asif RN, Ghadi YY, Ullah N, Khan MA, Al-Rasheed A (2024). YOLO and residual network for colorectal cancer cell detection and counting. Heliyon 10:e24403.

Hu J, Shen L, Albanie S, Sun G, Wu E (2018). Squeeze-and-excitation networks. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. /.

Huang ZJ, Patel B, Lu WH, Yang TY, Tung WC, Bucinskas V, Greitans M, Wu YW, Lin PT (2023). Yeast cell detection using fuzzy automatic contrast enhancement (FACE) and you only look once (YOLO). Sci Rep 13:16222.

Ito A, Hagiyama M, Oonuma J (2008). Nerve–mast cell and smooth muscle–mast cell interaction mediated by cell adhesion molecule-1, CADM1. J Smooth Muscle Res 44:83–93.

Johnson MM, Mendoza R, Raghavendra AJ, Podila R, Brown JM (2017). Contribution of engineered nanomaterials physicochemical properties to mast cell degranulation. Sci Rep 7:43570.

Kahveci B, Onen S, Akal F, Korkusuz P (2023). Detection of spermatogonial stem/progenitor cells in prepubertal mouse testis with deep learning. J Assist Reprod Genet 40:1187–95.

Khanam R, Hussain M (2024). YOLOv11: An overview of the key architectural enhancements. arXiv preprint abs/2410.17725:/.

Klatt EC, Lukes RJ, Meyer PR (1983). Benign and malignant mast-cell proliferations – diagnosis and separation using a pH-dependent toluidine blue stain in tissue section. Cancer 51:1119–24.

Krystel-Whittemore M, Dileepan KN, Wood JG (2016). Mast cell: A multi-functional master cell. Front Immunol 6:620.

Lin C, Yang P, Wang Q, Qiu Z, Lv W, Wang Z (2023). Efficient and accurate compound scaling for convolutional neural networks. Neural Netw 167:787–97.

Reber LL, Sibilano R, Starkl P, Roers A, Grimbaldeston MA, Tsai M, Gaudenzio N, Galli SJ (2017). Imaging protective mast cells in living mice during severe contact hypersensitivity. JCI Insight 2:e92900.

Sazak H, Kotan M (2025). Automated blood cell detection and classification in microscopic images using YOLOv11 and optimized weights. Diagnostics 15:22.

Shi CY, Zhu DL, Zhou CJ, Cheng S, Zou CY (2024). Gpmb-YOLO: a lightweight model for efficient blood cell detection in medical imaging. Health Inf Sci Syst 12:24.

Srebro D, Dozic B, Vuckovic S, Vujovic KS, Brkic BM, Dozic I, Srebro M (2023). The interactions of magnesium sulfate and cromoglycate in a rat model of orofacial pain; the role of magnesium on mast cell degranulation in neuroinflammation. Int J Mol Sci 24:6241.

Sun H, Wan XR, Tang SG, Li YN (2025). SSW-YOLO: Enhanced blood cell detection with improved feature extraction and multi-scale attention. J Imaging Inform Med /:01460–3.

Tan M, Le QV (2021). EfficientNetV2: Smaller models and faster training. In: International Conference on Machine Learning (ICML), vol. 139 of Proceedings of Machine Learning Research.

Vazquez TAS, Lopez NL, Carmona MCS (2024). MASTer cell: chief immune modulator and inductor of antimicrobial immune response. Front Immunol 15:1360296.

Wang X, Chen A, Pang Y, Hou C, Wang Y, Liu E, Zhao Y, Guo J, Li M (2025). Paeoniflorin as a potential agent for urticaria treatment: Suppressing mast cell degranulation through HMGB1/TLR4/NF-κB signaling inhibition. Mol Immunol 183:33–43.

Wang YX, Liu YH, Zhang ZL, Qiao X, Li YC, Ren LJ, Ding GH, Yao W, Yu Y (2024). Influence of acupuncture intensity on analgesic effects in AA rat models. Front Bioeng Biotechnol 12:1502535.

Weber A, Knop J, Maurer M (2001). Pattern recognition analysis of human cutaneous mast cell populations by topographical total body surface mapping. J Invest Dermatol 117:791–.

Wu J, Sun Y, Jiang Y, Bu Y, Chen C, Li J, Li L, Chen W, Cheng K, Xu J (2024). An automatic classification method of testicular histopathology based on SC-YOLO framework. Biotechniques 76:443–52.

Wu NK, Jia DY, Zhang CW, Li ZQ (2023). Cervical cell extraction network based on optimized YOLO. Math Biosci Eng 20:2364–81.

Wu Z, Shen C, van den Hengel A (2019). Wider or deeper: Revisiting the ResNet model for visual recognition. Pattern Recognit 90:119–33.

Yang BG, Kim AR, Lee D, An SB, Shim YA, Jang MH (2023). Degranulation of mast cells as a target for drug development. Cells 12:1506.

Zarnegar B, Mendez-Enriquez E, Westin A, Soderberg C, Dahlin JS, Gronvik KO, Hallgren J (2017). Influenza infection in mice induces accumulation of lung mast cells through the recruitment and maturation of mast cell progenitors. Front Immunol 8:310.

Zhong Z, Hou J, Yao Z, Dong L, Liu F, Yue J, Wu T, Zheng J, Ouyang G, Yang C, Song J (2024). Domain generalization enables general cancer cell annotation in single-cell and spatial transcriptomics. Nat Commun 15:1929.

Zhu MY, Yang P, Bian C, Zuo FF, Guo ZM, Wang YF, Wang YJ, Bai YX, Zhang N (2024). Convolutional neural network-assisted diagnosis of midpalatal suture maturation stage in cone-beam computed tomography. J Dent 141:104808.