3.2.7. Gynecological Cancer

A microfluidic chip for single-cell cultures utilizes self-assembled graphene oxide quantum dots (GOQDs) to facilitate high-activity single-cell cultures. This chip enables the maintenance of normal biomarker secretion in single cells and allows for efficient single cell separation at high throughputs. Consequently, it provides an ample amount of statistical data necessary for machine learning applications [120]. As a proof of concept, Wang et al. [120] developed a novel method for profiling single cells in real time using microfluidic chip technology and machine learning algorithms. They used this method to classify tumor cells based on the secreted biomarkers they produce. The microfluidic chip is designed to allow for the high-throughput analysis of single cells, enabling the measurement of multiple secreted biomarkers in real time. Then, machine learning algorithms were employed to analyze the data and classify the cells based on their biomarker profiles. The K-means strategy with machine learning was combined to analyze thousands of single tumor cell secretion data, resulting in the ability to classify tumor cells with a recognition accuracy of 95.0%.

As another example, Feng et al. [121] proposed the use of neural network-enhanced impedance flow cytometry (IFC) for the real-time, label-free, and non-invasive characterization of single cells based on intrinsic biophysical metrics. The method can obtain three intrinsic parameters (radius, cytoplasm conductivity, and specific membrane capacitance) online and in real time, achieving a significant improvement in the calculation speed. The experiments involved four cancer cell types and demonstrated a 91.5% classification accuracy. The paper suggests that this method could provide a new platform for high-throughput, real-time, and online cell intrinsic electrical characterization. Table 2 summarizes the electrical-based biosensors in conjunction with machine learning for cancer detection.


**Table 2.** Comparison of different electrical-based biosensors with ML analysis for cancer cell detection.
