Exosomes, known as nanometer‐sized vesicles (30–200 nm), are secreted by many types of cells. Cancer‐derived exosomes have great potential to be biomarkers for early clinical diagnosis and evaluation of cancer therapeutic efficacy. Conventional detection methods are limited to low sensitivity and reproducibility. There are hundreds of papers published with different detection methods in recent years to address these challenges. Therefore, in this review, pioneering researches about various detection strategies are comprehensively summarized and the analytical performance of these tests is evaluated. Furthermore, the exosome molecular composition (protein and nucleic acid) profiling, a single exosome profiling, and their application in clinical cancer diagnosis are reviewed. Finally, the principles and applications of machine learning method in exosomes researches are presented.

Airborne pathogens have been considered as highly infectious and transmittable between humans. With the pandemic outbreak of the coronavirus disease 2019 (COVID-19), an on-site diagnostic systemintegrated airborne pathogen-monitoring machine is recommended by experts for preventing and controlling the early stage β-coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spread. In this work, a small-volume rotating microfluidic fluorescence chip-integrated aerosol SARS-CoV-2 sampling system was constructed to satisfy the demand for rapid on-site sample collection and detection of SARS-CoV-2. The rotating microfluidic fluorescence system with small volume has very high sensitivity in the detection of SARS-CoV-2 (detection limit of 10 copies/μL with the shortest Ct value of 15 min), which is comparable to reverse transcription polymerase chain reaction (RT-PCR). The precision variation coefficients within/between batches are very low [coefficient of variation (CV) % ≤ 5.0%]. Our work has passed the comprehensive inspection of the microfluidic chip performance by the Shanghai Medical Device Testing Institute [National Medical Inspection (Design) no. 4408] and successfully tested 115 clinical samples. The integrated system exhibits 100% specificity, high sensitivity (10 copies/μL), and good precision (CV % ≤ 5.0%) in the rapid detection of SARS-CoV-2, thus realizing rapid monitoring and diagnostics of SARS-CoV-2 nucleic acid on-site.

Polymer dots (Pdots) have become attractive electrochemiluminescence (ECL) luminophores due to their facile synthesis, easy modification, and stable electrochemical and optical properties. However, their ECL efficiency is not high enough for practical applications. In this work, we proposed an ECL immunosensor based on localized surface plasmon resonance (LSPR) between AuNPs and Pdots for the determination of pancreatic cancer exosomes. Based on the finite-difference time-domain simulations and the band energy of Pdots and AuNPs, we proposed the possible LSPR mechanism. The hot electrons of plasmonic AuNPs were photoexcited to surface plasmon states by ECL emission of Pdots, and then the excited hot electrons were transferred to the conduction band of Pdots, which significantly improved the ECL efficiency of Pdots. The ECL immunosensor displayed a wide calibration range of 1.0 × 10 3 to 1.0 × 10 6 particles/mL with a detection limit of 400 particles/mL. Cancer-related protein profiling revealed high selectivity toward different expressions of exosomal surface proteins from PANC-01, HeLa, MCF-7, and HPDE6-C7 cell lines. The proposed ECL system exhibits a promising prospect for protein biomarker profiling and early cancerrelated diagnosis.