**4. Conclusions**

The pressing need to identify cancer in its earliest stages while avoiding invasive treatments has spurred the integration of innovative sensory techniques with cutting-edge machine learning algorithms. This fusion holds the potential to create a future where individuals can conveniently and promptly detect cancer within the confines of their homes. With advancements in detection technology and machine learning algorithms, our aim is to detect cancer at its early stage. Several studies have achieved highly accurate results in excess of 90% with optical biosensors, regardless of the type of cancer cell being detected or the ML algorithm used in the study. Among all the research papers analyzed for this study, most teams utilized ANNs for the machine learning aspect of their optical detection setups. On the other hand, some studies using electrical biosensors achieved slightly lower, yet consistently high, results when compared to the teams that employed optical biosensors. Slightly more than half of these teams recorded an accuracy greater than 90%, while the remaining teams had accuracies that were slightly lower. Most of these teams used SVMs to incorporate machine learning into their research, with ANNs being used to a lesser extent than in the optical detection teams. With further progress and advancements in these methodologies, we can hope for continuous improvements in the results and eventually strive towards a cancer-free future.

Biosensors are analytical devices that combine biological components, known as bioreceptors, with transducers to detect specific biological or chemical analytes. Despite the significant advancements, biosensors still face challenges related to the bioreceptor immobilization matrices, immobilization efficiency, and predicting responses in complex matrices. Machine learning (ML) can play a vital role in addressing these issues. For instance, ML models can assist in selecting the most suitable immobilization matrix for a specific bioreceptor by considering factors such as the bioreceptor type, analyte characteristics, and environmental conditions. This predictive capability helps researchers optimize the immobilization process and anticipate and correct deviations in sensor responses. Additionally, ML can aid in sensor calibration and data fusion, enhancing the accuracy and reliability of biosensor readings by continuously monitoring and adjusting the sensor responses based on historical data and real-time measurements.

In summary, biosensors are essential analytical tools with some inherent limitations. ML can offer valuable solutions by assisting in the selection of immobilization matrices for bioreceptors and improving sensor calibration and data fusion processes. These ML-driven interventions enhance the overall performance and reliability of biosensors, making them more effective in applications such as cancer cell detection and other complex analytical tasks.

**Author Contributions:** Conceptualization, M.K. and M.N.T.; methodology, M.K., D.S. and M.N.T.; validation, M.K., M.N.T. and D.S.; writing—original draft preparation, M.K.; writing—review and editing, M.N.T., D.S. and M.K.; supervision, M.J.; project administration, M.J.; funding acquisition, M.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Science Foundation grant numbers 1846740, and 2002511.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ethical constraints.

**Acknowledgments:** The authors are thankful to their respective institutions/universities for providing valuable support and funding to conduct this research work.

**Conflicts of Interest:** The authors have no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
