Multidomain Molecular Sensor Devices, Systems, and Algorithms for Improved Physiological Monitoring

Abstract

Molecular sensor systems, e.g., implantables and wearables, provide extensive health-related monitoring. Glucose sensor systems have historically prevailed in wearable bioanalysis applications due to their continuous and reliable glucose monitoring, a feat not yet accomplished for other biomarkers. However, the advancement of reagentless detection methodologies may facilitate the creation of molecular sensor systems for multiple analytes. Improving the sensitivity and selectivity of molecular sensor systems is also crucial for biomarker detection under intricate physiological circumstances. The term multidomain molecular sensor systems is utilized to refer, in general, to both biological and chemical sensor systems. This review examines methodologies for enhancing signal amplification, improving selectivity, and facilitating reagentless detection in multidomain molecular sensor devices. The review also analyzes the fundamental components of multidomain molecular sensor systems, including substrate materials, bodily fluids, power, and decision-making units. The review article further investigates how extensive data gathered from multidomain molecular sensor systems, in conjunction with current data processing algorithms, facilitate biomarker detection for precision medicine.

Keywords: biomarkers; healthcare physiological monitoring; machine learning; molecular sensors; precision medicine.

Figure 1

Chronological list of significant advancements in sensor systems’ development and biomarker identification facilitated by multidomain molecular sensor systems. (a) Advancements in molecular sensor systems for healthcare monitoring have been driven by data analysis, electronics, bioengineering, materials science, and telecommunication technologies [1,14,15,16,17,18,19,20,21,22,23]. The growing interest in healthcare physiological monitoring has led to cost reductions, enabling widespread adoption of these sensor systems globally. This has enabled continuous monitoring on a large scale. Additionally, advancements in fabrication techniques have allowed for increased complexity at smaller sizes, enabling molecular sensor systems to be integrated into personal technologies. (b) Precision medicine utilizes traditional blood or tissue biopsy techniques for biomarker detection, and the process for creating multidomain molecular sensor systems to detect biomarkers. Figure created with BioRender (BioRender.com).

Figure 2

The main methodologies for signal transduction and recognition in different molecular sensor devices. (a) Methods for enhancing biological-based sensor devices. DNAzymes and nanozymes offer alternate approaches for monitoring analytes through catalytic processes. Synthetic biomarkers can be utilized to improve molecular activities and aid in their identification within the body. Intracellular sensing facilitates the identification of novel, high-concentration biomarkers. (b) Signal transduction and recognition for chemical-based sensor devices. Voltammetric sensor devices detect reduction or oxidation in an electroactive molecule, e.g., uric acid, by allowing electrons to be received on or released from the electrode’s surface, with the observed peak current magnitude directly proportional to the target analyte’s concentration. Colorimetric sensor devices utilize a substrate to immobilize a color indicator molecule, causing a color change when the analyte comes into contact. Horseradish peroxidase (HRP) detects hydrogen by altering the redox state of chromophore molecule 3,3’,5,5’-teraenthykbenzidine (TMB) during H₂O₂ enzymatic breakdown. The intensity of color change or light absorption corresponds to the substance being analyzed. Potentiometric ion-selective sensor devices utilize a working electrode with an ion-selective membrane, containing ionosphere molecules, e.g., valinomycin, which interacts with specific ions of specific size and charge, resulting in a potential that varies the target ion concentration. Figure created with BioRender (BioRender.com).

Figure 3

Summary of multidomain sensor system’s fundamental components and multidomain molecular sensors’ data-driven procedure and algorithm overview. (a) Body-based molecular sensor systems consist of substrate and electrode materials, sensing devices, bodily fluids, power and decision-making units, which are the fundamental components. (b) The principles of three exemplary supervised and unsupervised learning algorithms. Schematic illustrations of support vector machines, linear discriminant analysis, and random forest algorithms for addressing classification problems. Schematic representations of principal component analysis, t-SNE clustering techniques for the reduction in data dimensionality. Schematic diagram of k-means clustering for data cluster analysis. Figure created with BioRender (BioRender.com).