Page 53 - Read Online

Page 32 of 44 Jung et al. Soft Sci 2024;4:15 https://dx.doi.org/10.20517/ss.2024.02

For the advancement of DM monitoring, signal analysis using ML or deep learning techniques has been
extensively researched for purposes such as forecasting glycemia, hypoglycemia, predictive monitoring for
glucose levels, and insulin recommendation. Currently, these studies mainly utilize electrochemical signals
derived from a single biomarker collected via invasive or minimally invasive methods, such as blood glucose
or commercial CGM devices, as inputs for model training. Although there is research on applying the
decision tree (DT) model to data non-invasively collected from wearable sweat-based glucose sensors to
measure actual glucose concentration from sweat, this approach still faces limitations in continuous and
predictive monitoring due to insufficient samples and unreliable concentration matching. For reliable,
accurate, and timely monitoring of DM using wearable electrochemical sensors, developing new artificial
intelligence (AI) models is essential. These models should be capable of matching the concentrations of
biomarkers from various biofluids to their actual concentrations, quantitatively comparing the
concentration differences between non-diabetics and diabetics, analyzing the correlations among multi-
biomarkers, and examining the trends in concentrations according to the lifestyle and dietary habits of an
individual.

CONCLUSIONS AND OUTLOOKS
In summary, this comprehensive review provides a thorough analysis of wearable sensor technology
employing electrochemical methodologies for managing DM and its associated complications. The focal
point is on developing non-invasive and minimally invasive electrochemical wearable sensors tailored for
detecting diverse metabolites and electrolytes within human biofluids. The review encompasses a detailed
exploration of electrochemical sensing mechanisms, the array of materials employed - including enzymes,
non-enzyme assays, and polymer-based ISMs - and the diverse sensing modalities integrated into wearable
electrochemical sensors. An in-depth examination of device architectures designed for monitoring specific
metabolites and electrolytes in biofluids, such as sweat, tears, saliva, and ISF, is undertaken to augment the
DM management efficacy. Furthermore, the review delves into wearable sensor devices engineered for a
multiplexed monitoring platform, encompassing chemical-electrophysiological hybrid sensing systems and
multiplexed electrochemical sensors capable of simultaneous monitoring of various biomarkers to enhance
the DM management precision. The discourse extends to ML-based multiplexed analysis methodologies
strategically devised for the proficient management of DM and its associated complications.

The escalating prevalence of DM has accelerated a growing reliance on glucose monitoring devices,
necessitating the revision of care guidelines for diabetic individuals. Researchers have explored diverse
diagnostic approaches beyond traditional glucose-related electrochemical reactions for DM management.
Among these advancements are the incorporation of novel nanomaterial reaction pathways and enzyme
electrodes and the merging of non-invasive or minimally invasive electrochemical sensor platforms with
wearable bioelectronics and self-powered biosensors. While facing hurdles in advanced material technology
and computational modeling, including accuracy, reliability, and environmental sustainability issues,
endeavors are ongoing to create sensor systems with improved precision. Recognizing the limitations of
single biomarker detection, there is a growing emphasis on multi-biomarker analysis technology. The
convergence of technologies is emerging as a trend, facilitating the integration of different surface chemical
compositions and detection principles into a single device. Techniques such as principal component
analysis are employed to simplify the complexity of high-dimensional data, enhancing the maturity of
detecting various combinations of biochemical indices and mitigating the impact of conflicting indicators
on results. Future strategies for diabetes-related sensing prioritize improvements in cost-effectiveness,
accuracy, precision, selectivity, and stability, aiming to provide more comfortable and secure conditions for
patients. Advancements in miniaturization, portability, and enhanced software and hardware facilities have
been made through various research studies [Table 8], and these advancements are anticipated in the