基于深度学习的APL相关嗜天青颗粒阳性显微图像自动判别方法

doi:10.3969/j.issn.0258-8021.2026.03.004

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (5699 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract The aim of this work is to develop a deep learning method for automatically identifying acute promyelocytic leukemia (APL)-related azurophilic granule-positive microscopic images, to improve the efficiency of morphological prescreening and support the early auxiliary diagnosis of APL in emergency clinical settings. A total of 450 leukocyte microscopic images collected from Nanhai People′s Hospital were used to construct an image-level binary classification dataset for the presence or absence of typical azurophilic granules, including 230 negative and 220 positive images. The dataset was divided into training and test sets at a ratio of 8∶2 using stratified random sampling. An end-to-end network, termed EADF-Net, was proposed based on EfficientNet-B4, which incorporated a feature pyramid attention module to capture multi-scale contextual information, a multi-scale feature fusion module to enhance fine-grained granule and boundary representations, and a context enhancement module to aggregate local and global contextual features. During training, binary cross-entropy loss, a Sobel gradient-based boundary enhancement regularization term, and supervised contrastive loss were jointly optimized. The proposed model was compared with ResNet-50, DenseNet-121, EfficientNet variants, and Swin Transformer under the same training settings. On the test set, EADF-Net achieved an AUC of 92.2%, accuracy of 90.0%, sensitivity of 84.1%, specificity of 95.7%, and F1-score of 89.1%. Compared with ResNet-50, EfficientNet-B4, EfficientNetV2-S, and Swin Transformer, EADF-Net showed better overall performance while maintaining a smaller model size. Ablation experiments further demonstrated that the combination of FPA, MSFF, and CEM effectively improved the model′s ability to discriminate fine-grained morphological features of APL. In conclusion, EADF-Net enabled stable discrimination of APL-related azurophilic granule-positive microscopic images under limited-sample conditions, while achieving improved performance with controlled model complexity, suggesting its potential as an auxiliary tool for early APL screening.

Key words： acute promyelocytic leukemia deep learning azurophilic granules microscopic image classification multi-scale feature fusion

Received: 10 December 2025

PACS:

R318

Corresponding Authors: ^*E-mail:xujiayu0130@Hotmail.com

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Lin Zhiyuan
	Liu Xin
	Qiu Jinming
	Mai Xiuqu
	Zhang Fuhua
	Xu Jiayu

Cite this article:

Lin Zhiyuan,Liu Xin,Qiu Jinming, et al. Deep Learning-Based Image-Level Detection of APL-Related Azurophilic Granule-PositiveLeukemia Cells[J]. Chinese Journal of Biomedical Engineering, 2026, 45(3): 288-298.

URL:

http://cjbme.csbme.org/EN/10.3969/j.issn.0258-8021.2026.03.004 OR http://cjbme.csbme.org/EN/Y2026/V45/I3/288

[1] Gǎman MA, Dug ǎeşescu M, Popescu DC. Applications of artificial intelligence in acute promyelocytic leukemia: an avenue of opportunities? a systematic review [J]. Journal of Clinical Medicine, 2025; 14(5):1670.
[2] Bryan C. Hambley, Kelly J. Norsworthy, Jagar Jasem, et al. Fibrinogen consumption and use of heparin are risk factors for delayed bleeding during acute promyelocytic leukemia induction [J]. Leukemia Rresearch, 2019, 83: 106174.
[3] Lo-Coco F, Avvisati G, Vignetti M, et al. Retinoic acid and arsenic trioxide for acute promyelocytic leukemia [J]. New England Journal of Medicine, 2013, 369(2): 111-121.
[4] Sanz MA, Fenaux P, Tallman MS, et al. Management of acute promyelocytic leukemia: updated recommendations from an expert panel of the European LeukemiaNet [J]. Blood, 2019, 133(15): 1630-1643.
[5] 江梅. 急性早幼粒细胞白血病临床特征及罕见融合基因CPSF6-RARG的致病作用研究 [D]. 南昌大学, 2021.
[6] Borse S, Wang Y, Zhang Y, et al. Inverseform: a loss function for structured boundary-aware segmentation [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, 2021: 5901-5911.
[7] Sidhom JW, Siddarthan IJ, Lai BS, et al. Deep learning for diagnosis of acute promyelocytic leukemia via recognition of genomically imprinted morphologic features[J]. npj Precis, 2021, 5: 38.
[8] Manescu P, Narayanan P, Bendkowski C, et al. Detection of acute promyelocytic leukemia in peripheral blood and bone marrow with annotation-free deep learning[J]. Sci Rep, 2023, 13: 2562.
[9] Eckardt JN, Middeke JM, Riechert S, et al. Deep learning detects acute myeloid leukemia and predicts NPM1 mutation status from bone marrow smears [J]. Leukemia, 2022, 36(1): 111-118.
[10] Glüge S, Balabanov S, Koelzer VH, et al. Evaluation of deep learning training strategies for the classification of bone marrow cell images[J]. Computer Methods and Programs in Biomedicine, 2024, 243: 107924.
[11] Wang T, Lu J, Lai Z, et al. Uncertainty-guided pixel contrastive learning for semi-supervised medical image segmentation. IJCAI, 2022: 1444-1450.
[12] Srisuwananukorn A, Salama ME, Pearson AT. Deep learning applications in visual data for benign and malignant hematologic conditions: a systematic review and visual glossary[J]. Haematologica, 2023, 108(8): 1993.
[13] Sharmay Y, Ehsany L, Syed S, et al. HistoTransfer: understanding transfer learning for histopathology. IEEE, 2021: 1-4.
[14] Mingxing Tan, Quoc V. Le. EfficientNetV2: Smaller Models and Faster Training. ICML 2021: 10096-10106.
[15] Ouyang N, Wang W, Ma L, et al. Diagnosing acute promyelocytic leukemia by using convolutional neural network[J]. Clin Chim Acta, 2021, 512: 1-6.
[16] Qiao Y, Zhang Y, Liu N, et al. An end-to-end pipeline for early diagnosis of acute promyelocytic leukemia based on a compact CNN model[J]. Diagnostics, 2021, 11(7): 1237.
[17] Boldú L, Merino A, Acevedo A, et al. A deep learning model (ALNet) for the diagnosis of acute leukaemia lineage using peripheral blood cell images[J]. Computer Methods and Programs in Biomedicine, 2021, 202: 105999.
[18] Tan Mingxing,Le QV. EfficientNetV2: smaller models and faster training[C]//Proceedings of the 38^th International Conference on Medicine Learning. Vienna: ICML 2021: 10096-10106.
[19] Oybek Kizi RF, Theodore Armand TP, Kim HC. A review of deep learning techniques for leukemia cancer classification based on blood smear images[J]. Applied Biosciences, 2025, 4(1): 9.
[20] Yu Z, Li J, Wen X, et al. RETRACTED ARTICLE: AMLnet, A deep-learning pipeline for the differential diagnosis of acute myeloid leukemia from bone marrow smears[J]. Journal of hematology & oncology, 2023, 16(1): 27.
[21] Abd El-Aziz AA, Mahmood MA, Abd El-Ghany S. A robust EfficientNetV2-S classifier for predicting acute lymphoblastic leukemia based on cross validation. symmetry, 2024, 17(1): 24.
[22] Wang X, Pan G, Hu Z, et al. A two stage blood cell detection and classification algorithm based on improved YOLOv7 and EfficientNetv2[J]. Sci Rep, 2025, 15: 8427.
[23] Abdelrahman A, Viriri S. EfficientNet family U-Net models for deep learning semantic segmentation of kidney tumors on CT images[J]. Frontiers in Computer Science, 2023, 5: 1235622.
[24] Batool A, Byun YC. Lightweight EfficientNetB3 model based on depthwise separable convolutions for enhancing classification of leukemia white blood cell images[J]. IEEE Access, 2023, 11: 37203-37215.
[25] 中华医学会血液学分会, 中国医师协会血液科医师分会. 中国急性早幼粒细胞白血病诊疗指南(2018年版)[J]. 中华血液学杂志, 2018, 39(3):179-183.
[26] Eckardt JN, Schmittmann T, Riechert S, et al. Deep learning identifies acute promyelocytic leukemia in bone marrow smears[J]. BMC Cancer, 2022, 22(1): 201.