Classification between Normal and Cancerous Human Urothelial Cells by Using Micro-Dimensional Electrochemical...

Area of Science:

• Biomedical Engineering
• Oncology
• Analytical Chemistry

Background:

• Urothelial cancer of the bladder (UCB) presents high incidence and recurrence rates, posing a significant health burden.
• Current diagnostic methods like voided urine cytology have limited sensitivity, necessitating improved noninvasive detection tools.
• There is a critical need for accurate, noninvasive methods for early UCB detection and monitoring.

Purpose of the Study:

• To develop an intelligent, noninvasive diagnostic tool for distinguishing normal urothelial cells from UCB cells.
• To leverage micro-electrochemical impedance spectroscopy (µEIS) combined with machine learning (ML) for high-accuracy UCB detection.
• To create a sensitive and reliable alternative to existing low-sensitivity diagnostic techniques for UCB.

Main Methods:

• A unique valved flow cytometry system equipped with a pneumatic valve was developed to enhance sensitivity and prevent cell clogging.
• Micro-dimensional electrochemical impedance spectroscopy (µEIS) was employed to measure impedance spectra of urothelial cells at various frequencies (10 kHz to 1 MHz).
• Six machine learning (ML) models were trained and compared using 236 impedance spectra, with hyperparameters optimized via grid search or Bayesian optimization.

Main Results:

• The µEIS sensor effectively discriminated between normal (SV-HUC-1) and cancerous (TCCSUP) urothelial cells due to confined electric fields.
• The random forest ML model demonstrated superior performance, achieving 91.7% accuracy, 92.9% sensitivity, 92.9% precision, 90% specificity, and a 93.8% F1-score.
• The developed system showed high potential for sensitive and accurate UCB cell discrimination.

Conclusions:

• The combination of µEIS and ML provides a promising noninvasive approach for UCB detection.
• This intelligent discrimination method offers a significant advancement over traditional low-sensitivity diagnostic techniques.
• Further development could lead to a clinically viable tool for early UCB diagnosis and patient management.