GhostCell-Net: A Lightweight Deep Learning Framework with Mobile Deployment for Automated Classification of Arsenic-Exposed Cells
DOI:
https://doi.org/10.4108/eetiot.14885Keywords:
Lightweight Deep Learning, Mobile Deployment, Arsenic Exposure Detection, Attention Mechanism, Cellular Image ClassificationAbstract
INTRODUCTION: Chronic arsenic exposure from contaminated drinking water remains a critical public health concern, particularly in regions that rely on private wells. Accurate assessment of the impact of arsenic on cellular morphology requires computational tools that are both effective and accessible outside centralised laboratories.
OBJECTIVES: This work aims to develop a lightweight yet accurate deep learning model for the automated classification of arsenic-exposed PC12 cells, and to deliver it as a complete, on-device diagnostic system rather than an offline research prototype.
METHODS: We propose GhostCell-Net, which builds upon the MobileNetV2 backbone and introduces three plug-in enhancement modules: Multi-Scale Dilated Aggregation (MSDA) for capturing cell features across multiple spatial extents, Frequency-Channel Attention (FCA) for texture-aware channel recalibration using DCT-inspired frequency descriptors, and a Spatial Gating Unit (SGU) for efficient spatial attention via channel-split gating. These modules are combined through a Progressive Stage Fusion (PSF) mechanism that aggregates multi-granularity features for classification, and the network is trained with a two-phase schedule using differential learning rates. The trained model is exported to PyTorch Mobile and integrated into an Android application supporting real-time camera capture, batch image processing, multi-model switching and automated accuracy evaluation.
RESULTS: On the Arsenic Image Dataset, GhostCell-Net reaches 93.162% test accuracy with approximately 2.7M parameters, improving on the attention-based CNN baseline by 4.542 percentage points while using roughly 8.7 times fewer parameters. In on-device evaluation it attains 86.27% accuracy in 1567 ms, against 84.31% in 73068 ms for the ResNet-CBAM baseline.
CONCLUSION: GhostCell-Net achieves competitive classification accuracy at a substantially reduced computational cost and demonstrates a practical path from model design to point-of-care mobile deployment for cellular image analysis
Downloads
References
[1] Maarefdoust, R., Currie, D., MacLeod, B., Song, Y., Concannon, J., Frankel, S. and Zhang, X. (2024) Attention-based cnn for enhanced detection of arsenic exposure. In International Conference on Medical Imaging and Computer-Aided Diagnosis (Springer): 55–65.
[2] Wasserman, G.A., Liu, X., Parvez, F., Ahsan, H., Factor-Litvak, P., van Geen, A., Slavkovich, V. et al. (2004) Water arsenic exposure and children’s intellectual function in araihazar, bangladesh. Environmental health perspectives 112(13): 1329–1333.
[3] Wasserman, G.A., Liu, X., Parvez, F., Ahsan, H., Factor- Litvak, P., Kline, J., Van Geen, A. et al. (2007) Water arsenic exposure and intellectual function in 6-year-old children in araihazar, bangladesh. Environmental health perspectives 115(2): 285–289.
[4] Wasserman, G.A., Liu, X., LoIacono, N.J., Kline, J., Factor-Litvak, P., van Geen, A., Mey, J.L. et al. (2014) A cross-sectional study of well water arsenic and child iq in maine schoolchildren. Environmental Health 13: 1–10.
[5] Frankel, S., Concannon, J., Brusky, K., Pietrowicz, E., Giorgianni, S., Thompson, W.D. and Currie, D.A. (2009) Arsenic exposure disrupts neurite growth and complexity in vitro. Neurotoxicology 30(4): 529–537.
[6] Rajpurkar, P., Irvin, J., Ball, R.L., Zhu, K., Yang, B., Mehta, H., Duan, T. et al. (2018) Deep learning for chest radiograph diagnosis: A retrospective comparison of the chexnext algorithm to practicing radiologists. PLoS medicine 15(11): e1002686.
[7] Song, Y., Siriwardane, E.M.D., Zhao, Y. and Hu, J. (2021) Computational discovery of new 2d materials using deep learning generative models. ACS Applied Materials & Interfaces 13(45): 53303–53313.
[8] Fletcher, S., Scott, G., Currie, D., Zhang, X., Song, Y. and MacLeod, B. (2025) Deep learning for taxol exposure analysis: a new cell image dataset and attention-based baseline model. arXiv preprint arXiv:2508.14349 .
[9] Fletcher, S., Scott, G., Currie, D., Zhang, X., Song, Y. and MacLeod, B. (2025) Interpretable tile-based classification of paclitaxel exposure. In International Symposium on Visual Computing (Springer): 361–372.
[10] Shafiq, M. and Gu, Z. (2022) Deep residual learning for image recognition: A survey. Applied sciences 12(18): 8972.
[11] Muneeb, M., Feng, S.F. and Henschel, A. (2022) Deep learning pipeline for image classification on mobile phones. arXiv preprint arXiv:2206.00105 .
[12] Kornilova, A., Salnikov, M., Novitskaya, O., Begicheva, M., Sevriugov, E., Shcherbakov, K., Pronina, V. et al. (2021) Deep learning framework for mobile microscopy. In 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) (IEEE): 324– 328.
[13] Sandler, M., Howard, A., Zhu, M., Zhmoginov, A. and Chen, L.C. (2018) Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition: 4510–4520.
[14] He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition: 770–778.
[15] Nagao, Y., Sakamoto, M., Chinen, T., Okada, Y. and Takao, D. (2020) Robust classification of cell cycle phase and biological feature extraction by image-based deep learning. Molecular biology of the cell 31(13): 1346–1354.
[16] Kumar, S. and Dawani, A. (2024) Deep learning for radiology: Improving diagnostic accuracy in medical imaging. Spectrum of Engineering Sciences : 517–533.
[17] Jin, X., Zou, Y. and Huang, Z. (2021) An imbalanced image classification method for the cell cycle phase. Information 12(6): 249.
[18] Hu, J., Shen, L. and Sun, G. (2018) Squeeze-and- excitation networks. In Proceedings of the IEEE conference on computer vision and pattern recognition: 7132–7141.
[19] Woo, S., Park, J., Lee, J.Y. and Kweon, I.S. (2018) Cbam: Convolutional block attention module. In Proceedings of the European conference on computer vision (ECCV): 3–19.
[20] Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M. et al. (2016) Tensorflow: a system for large-scale machine learning. In 12th USENIX symposium on operating systems design and implementation (OSDI 16): 265–283.
[21] Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T. et al. (2019) Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems 32.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Yunlong Li, Yuqi Song, Xin Zhang, Chuxiong Wu
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution CC BY 4.0 license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work is properly cited.
