$\mu$MobileScan: A Smartphone Whole Slide Imaging App for Red Blood Cell Counting with Real-Time Guidance

This repository includes the Android app implementation of our paper.

Abstract

Red blood cell counting is an essential metric in hematology analysis for disease diagnosis. Low red blood cell count can indicate anemia, and high red blood cell count can indicate tumors. The current clinical standard for counting the cells involves human experts counting the cells on hemocytometers through a microscope. This approach is labor-intensive and volatile. The manual process of counting cells via hemocytometer has no record (i.e., images) other than the clinicians' count. Whole slide imaging (WSI) systems can be used for slide scanning and digitization, but the high cost limits access to WSI technology. We propose a smartphone app called $\mu$MobileScan to make this measure more readily available. $\mu$MobileScan solely relies on commonly available tools, smartphone cameras, and microscopes, to reconstruct a virtual WSI hemocytometer by providing real-time guidance to the users and operating cell counting on the constructed images. We conducted studies to evaluate the WSI construction quality, learning curve, and cell counting accuracy demonstrating usability. The studies demonstrated the impact of scanning routes and overlap percentage on feature mapping. The learning curve evaluation on 8 participants illustrated that our system can guide users to scan the hemocytometer with consistent overlap percentages, and the scanning time and success rate converge after 5 trials. Usability testing with four participants (two experts and two novices) on de-identified samples from three donors under various configurations achieved high cell counting accuracy (MAPE = 1.40%, SDAPE = 1.48%). Evaluations across diverse hardware and optical configurations demonstrated the robustness and practicality of the proposed system.

If you refer to this work or the app in your research, please cite our paper:

@article{10.1145/3790109,
  author = {Wu, Sixuan and Bae, Jung Hyun and Adams, Alexander T.},
  title = {μMobileScan: A Smartphone Whole Slide Imaging App for Red Blood Cell Counting with Real-time Guidance},
  year = {2026},
  issue_date = {March 2026},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {10},
  number = {1},
  url = {https://doi.org/10.1145/3790109},
  doi = {10.1145/3790109},
  month = mar,
  articleno = {22},
  numpages = {28},
}