AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics

Hanqi Hu; Zongliang Guo; Siyu Hu; Mengjie Li; Tao Zhao; Hang Li; Kangfu Chen; Shuailong Zhang; Haixia Li; Rongxin Fu

doi:10.1109/MEMS64181.2026.11419286

AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics

Hanqi Hu
, Zongliang Guo
, Siyu Hu
, Mengjie Li
, Tao Zhao
, Hang Li
, Kangfu Chen
, Shuailong Zhang^*
, Haixia Li^*
, Rongxin Fu^*

^*Corresponding author for this work

Beijing Institute of Technology
CAS - Suzhou Institute of Biomedical Engineering and Technology
Peking University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Bladder cancer poses a significant global health challenge due to its high recurrence rate and the limitations of current diagnostic methods. While cystoscopy remains the clinical gold standard, it is invasive and costly, whereas non-invasive alternatives such as urine cytology suffer from low sensitivity, especially in early-stage tumors. Therefore, this paper presents a label-free platform integrating hyperspectral imaging (HSI) with digital microfluidics to overcome the reliance on biomarkers. A deep learning model was implemented to achieve accurate sorting of exfoliated tumor cells in urine(UTCs) from endothelial progenitor cells (EPCs). Biophysical validation confirmed distinct nanoscale structural heterogeneity in UTCs, characterized by local refractive index variations and spatial disorder, demonstrating the platform's unique capability to leverage intrinsic biophysical properties for cancer cell identification without the need for labeling.

Original language	English
Title of host publication	2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	580-583
Number of pages	4
ISBN (Electronic)	9798331572518
DOIs	https://doi.org/10.1109/MEMS64181.2026.11419286
Publication status	Published - 2026
Event	39th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2026 - Salzburg, Austria Duration: 25 Jan 2026 → 29 Jan 2026

Publication series

Name	Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)
ISSN (Print)	1084-6999

Conference

Conference	39th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2026
Country/Territory	Austria
City	Salzburg
Period	25/01/26 → 29/01/26

Keywords

Cell Sorting
Deep Learning
Digital Microfluidics
Hyperspectral Computational Imaging

Access to Document

10.1109/MEMS64181.2026.11419286

Cite this

Hu, H., Guo, Z., Hu, S., Li, M., Zhao, T., Li, H., Chen, K., Zhang, S., Li, H., & Fu, R. (2026). AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics. In 2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026 (pp. 580-583). (Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MEMS64181.2026.11419286

Hu, Hanqi ; Guo, Zongliang ; Hu, Siyu et al. / AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics. 2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026. Institute of Electrical and Electronics Engineers Inc., 2026. pp. 580-583 (Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)).

@inproceedings{f384a96b6aa44f9ebf94392cb187c13d,

title = "AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics",

abstract = "Bladder cancer poses a significant global health challenge due to its high recurrence rate and the limitations of current diagnostic methods. While cystoscopy remains the clinical gold standard, it is invasive and costly, whereas non-invasive alternatives such as urine cytology suffer from low sensitivity, especially in early-stage tumors. Therefore, this paper presents a label-free platform integrating hyperspectral imaging (HSI) with digital microfluidics to overcome the reliance on biomarkers. A deep learning model was implemented to achieve accurate sorting of exfoliated tumor cells in urine(UTCs) from endothelial progenitor cells (EPCs). Biophysical validation confirmed distinct nanoscale structural heterogeneity in UTCs, characterized by local refractive index variations and spatial disorder, demonstrating the platform's unique capability to leverage intrinsic biophysical properties for cancer cell identification without the need for labeling.",

keywords = "Cell Sorting, Deep Learning, Digital Microfluidics, Hyperspectral Computational Imaging",

author = "Hanqi Hu and Zongliang Guo and Siyu Hu and Mengjie Li and Tao Zhao and Hang Li and Kangfu Chen and Shuailong Zhang and Haixia Li and Rongxin Fu",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 39th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2026 ; Conference date: 25-01-2026 Through 29-01-2026",

year = "2026",

doi = "10.1109/MEMS64181.2026.11419286",

language = "English",

series = "Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "580--583",

booktitle = "2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026",

address = "United States",

}

Hu, H, Guo, Z, Hu, S, Li, M, Zhao, T, Li, H, Chen, K, Zhang, S, Li, H & Fu, R 2026, AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics. in 2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026. Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS), Institute of Electrical and Electronics Engineers Inc., pp. 580-583, 39th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2026, Salzburg, Austria, 25/01/26. https://doi.org/10.1109/MEMS64181.2026.11419286

AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics. / Hu, Hanqi; Guo, Zongliang; Hu, Siyu et al.
2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026. Institute of Electrical and Electronics Engineers Inc., 2026. p. 580-583 (Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics

AU - Hu, Hanqi

AU - Guo, Zongliang

AU - Hu, Siyu

AU - Li, Mengjie

AU - Zhao, Tao

AU - Li, Hang

AU - Chen, Kangfu

AU - Zhang, Shuailong

AU - Li, Haixia

AU - Fu, Rongxin

PY - 2026

Y1 - 2026

N2 - Bladder cancer poses a significant global health challenge due to its high recurrence rate and the limitations of current diagnostic methods. While cystoscopy remains the clinical gold standard, it is invasive and costly, whereas non-invasive alternatives such as urine cytology suffer from low sensitivity, especially in early-stage tumors. Therefore, this paper presents a label-free platform integrating hyperspectral imaging (HSI) with digital microfluidics to overcome the reliance on biomarkers. A deep learning model was implemented to achieve accurate sorting of exfoliated tumor cells in urine(UTCs) from endothelial progenitor cells (EPCs). Biophysical validation confirmed distinct nanoscale structural heterogeneity in UTCs, characterized by local refractive index variations and spatial disorder, demonstrating the platform's unique capability to leverage intrinsic biophysical properties for cancer cell identification without the need for labeling.

AB - Bladder cancer poses a significant global health challenge due to its high recurrence rate and the limitations of current diagnostic methods. While cystoscopy remains the clinical gold standard, it is invasive and costly, whereas non-invasive alternatives such as urine cytology suffer from low sensitivity, especially in early-stage tumors. Therefore, this paper presents a label-free platform integrating hyperspectral imaging (HSI) with digital microfluidics to overcome the reliance on biomarkers. A deep learning model was implemented to achieve accurate sorting of exfoliated tumor cells in urine(UTCs) from endothelial progenitor cells (EPCs). Biophysical validation confirmed distinct nanoscale structural heterogeneity in UTCs, characterized by local refractive index variations and spatial disorder, demonstrating the platform's unique capability to leverage intrinsic biophysical properties for cancer cell identification without the need for labeling.

KW - Cell Sorting

KW - Deep Learning

KW - Digital Microfluidics

KW - Hyperspectral Computational Imaging

UR - https://www.scopus.com/pages/publications/105041737701

U2 - 10.1109/MEMS64181.2026.11419286

DO - 10.1109/MEMS64181.2026.11419286

M3 - Conference contribution

AN - SCOPUS:105041737701

T3 - Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)

SP - 580

EP - 583

BT - 2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 39th IEEE International Conference on Micro Electro Mechanical Systems, MEMS 2026

Y2 - 25 January 2026 through 29 January 2026

ER -

Hu H, Guo Z, Hu S, Li M, Zhao T, Li H et al. AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics. In 2026 IEEE 39th International Conference on Micro Electro Mechanical Systems, MEMS 2026. Institute of Electrical and Electronics Engineers Inc. 2026. p. 580-583. (Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS)). doi: 10.1109/MEMS64181.2026.11419286

AI-Driven Label-Free Exfoliated Tumor Cells Identification Using Hyperspectral Single-Cell Optofluidics

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this