Simulation-Based Bioelectronic Modeling for Clustering Cancer Cells by Malignancy in Biosensing Applications.

Area of Science:

• Biomedical Engineering
• Cancer Research
• Machine Learning Applications

Background:

• Organic electrochemical transistors (OECTs) offer label-free biosensing capabilities.
• Distinguishing cancer cell states, particularly metastatic potential, remains a challenge in diagnostics.
• Electrical impedance and frequency response analysis can reveal cellular characteristics.

Purpose of the Study:

• To develop and validate an OECT-based biosensing approach for cancer cell differentiation.
• To integrate unsupervised machine learning for analyzing OECT-derived electrical signatures.
• To correlate electrical properties with cellular metastatic potential.

Main Methods:

• Utilized organic electrochemical transistors (OECTs) to measure electrical responses of simulated cell lines.
• Generated a synthetic dataset capturing membrane capacitance, double-layer capacitance, and crossover frequency.
• Applied K-means clustering to identify distinct patterns in the electrical data.

Main Results:

• OECTs successfully captured unique electrical signatures differentiating cell states.
• Key electrical parameters (capacitance, crossover frequency) correlated with cell interaction and metastatic behavior.
• K-means clustering identified distinct cell groupings based on electrical properties.

Conclusions:

• OECTs are a promising technology for label-free cancer cell differentiation.
• Unsupervised clustering effectively maps electrical properties to metastatic potential.
• This approach demonstrates the potential for next-generation biosensing chips in cancer diagnostics.