MXenes and machine learning enable integrated multimodal biosensing of chemical and biological targets

초록

MXenes, a rapidly expanding class of two-dimensional transition-metal carbides and nitrides, have emerged as robust nanomaterials for biosensing applications due to their high electrical conductivity, tunable surface chemistry, broadband optical absorption, and efficient photothermal conversion. These intrinsic properties enable MXenes to engage in diverse transduction pathways simultaneously, facilitating the evolution of MXenebased biosensors from single-mode designs to sophisticated multimodal sensing systems. This review examines multimodal biosensing through a materials-focused lens and categorizes MXene-enabled platforms into homogeneous multimodal systems, which extract multiple signal states within a single physical domain, and heterogeneous multimodal systems, which integrate signals originating from distinct domains. While multimodal designs improve sensitivity, robustness, and self-validation, they also generate complex, high-dimensional datasets that can complicate conventional signal analysis. Machine learning enables multimodal biosensors to process complex, high-dimensional data, integrate multiple signals, and achieve accurate classification, quantitation, and optimized sensor design with improved efficiency and performance. Collectively, this review highlights recent developments at the intersection of MXenes, multimodal biosensing, and machine learning, and delineates prospects for advanced biosensing technologies.

키워드