Transforming cytokine diagnostics: AI, multiplexing, and point-of-care biosensing technologies
Authors:
- 1 November 2025
- journal article
- review article
- Published by Springer Nature in Microchimica Acta
- Vol. 192 (11) , 1-31
Abstract
Cytokines are central regulators of immune responses and have emerged as key biomarkers in diverse pathological conditions, including infections, autoimmune disorders, and cancer. Conventional laboratory methods for cytokine detection, while accurate, often lack the speed, portability, and multiplexing capacity required for timely clinical decision-making. Recent advances in biosensor technology—particularly at the point-of-care (POC)—are reshaping this landscape by enabling rapid, decentralized, and sensitive detection of cytokine panels in complex biological samples. AI-enabled multiplex POC platforms now achieve limits of detection as low as 0.01–100 pg/mL, with dynamic ranges spanning 3–4 orders of magnitude, using 1–50 µL of sample and delivering results within 5–30 min. Compared with centralized single-plex workflows, these systems provide faster, lower-volume, and more clinically actionable testing. Artificial intelligence further strengthens performance by providing calibrated predictive outputs, uncertainty estimates, and drift monitoring. This review highlights the convergence of multiplexed biosensing strategies with artificial intelligence (AI) to enhance the analytical performance, interpretability, and clinical utility of cytokine diagnostics. We first discuss the evolution from traditional platforms to portable and miniaturized systems, and then summarize emerging review literature addressing cytokine biosensing in contexts such as sepsis, metabolic disorders, and systemic inflammation. Next, we examine experimental studies demonstrating POC-compatible platforms for multiplexed cytokine detection, and finally focus on next-generation biosensors that integrate machine learning (ML) algorithms—including convolutional neural networks (CNNs) and decision-tree models—for autonomous signal processing and decision support. Despite challenges in validation, hardware integration, and explainability, these technologies hold transformative potential for real-time immune monitoring, precision medicine, and global health applications. Graphical abstractKeywords
This publication has 112 references indexed in Scilit:
- The role of cytokines in the pathogenesis of rheumatoid arthritis – Practical and potential application of cytokines as biomarkers and targets of personalized therapyCytokine, 2015
- A fully-automated, six-plex single molecule immunoassay for measuring cytokines in bloodJournal of Immunological Methods, 2015
- Multiplexed femtomolar quantitation of human cytokines in a fluoropolymer microcapillary filmThe Analyst, 2015
- Multiplex Serum Cytokine Immunoassay Using Nanoplasmonic Biosensor MicroarraysACS Nano, 2015
- A computational method to differentiate normal individuals, osteoarthritis and rheumatoid arthritis patients using serum biomarkersJournal of The Royal Society Interface, 2014
- Redirection of Cytokine Production By Lymphocytes From Women with Pre‐Term Delivery By DydrogesteroneAmerican Journal of Reproductive Immunology, 2007
- Validation and comparison of luminex multiplex cytokine analysis kits with ELISA: Determinations of a panel of nine cytokines in clinical sample culture supernatantsJournal of Reproductive Immunology, 2005
- Maternal Cytokine Production Patterns in Women with Pre‐eclampsiaAmerican Journal of Reproductive Immunology, 2005
- Increased Expression of Pro‐Inflammatory Cytokines in Placentas of Women Undergoing Spontaneous Preterm Delivery or Premature Rupture of MembranesAmerican Journal of Reproductive Immunology, 2004
- Cytokine production by maternal lymphocytes during normal human pregnancy and in unexplained recurrent spontaneous abortionHuman Reproduction, 2000