Machine learning and chemometrics for electrochemical sensors: moving forward to the future of analytical chemistry

Smartphone based wearable sweat glucose sensing device correlated with machine learning for real-time diabetes screening

BACKGROUND

Diabetes is a significant health threat, with its prevalence and burden increasing worldwide indicating its challenge for global healthcare management. To decrease the disease severity, the diabetic patients are recommended to regularly check their blood glucose levels. The conventional finger-pricking test possesses some drawbacks, including painfulness and infection risk. Nowadays, smartphone has become a part of our lives offering an important benefit in self-health monitoring. Thus, non-invasive wearable sweat glucose sensor connected with a smartphone readout is of interest for real-time glucose detection.

RESULTS

Wearable sweat glucose sensing device is fabricated for self-monitoring of diabetes. This device is designed as a body strap consisting of a sensing strip and a portable potentiostat connected with a smartphone readout via Bluetooth. The sensing strip is modified by carbon nanotubes (CNTs)-cellulose nanofibers (CNFs), followed by electrodeposition of Prussian blue. To preserve the activity of glucose oxidase (GOx) immobilized on the modified sensing strip, chitosan is coated on the top layer of the electrode strip. Herein, machine learning is implemented to correlate between the electrochemical results and the nanomaterial content along with deposition cycle of prussian blue, which provide the highest current response signal. The optimized regression models provide an insight, establishing a robust framework for design of high-performance glucose sensor.

SIGNIFICANCE

This wearable glucose sensing device connected with a smartphone readout offers a user-friendly platform for real-time sweat glucose monitoring. This device provides a linear range of 0.1-1.5 mM with a detection limit of 0.1 mM that is sufficient enough for distinguishing between normal and diabetes patient with a cut-off level of 0.3 mM. This platform might be an alternative tool for improving health management for diabetes patients.

Surface-engineered natural fibers: Emerging alternative substrates for chemical sensor applications: A review

Natural fiber has become one of the most widely used alternative materials for chemical sensor fabrication due to its advantages, such as biocompatibility, flexibility, and self-microfluidic properties. Enhanced natural fiber surface has been used as a substrate in colorimetric and electrochemical sensors. This review focuses on improving the natural fiber properties for preparation as a substrate for chemical sensors. Various methods for natural fiber extraction are discussed and compared. Bleaching and decolorization is important for preparation of colorimetric sensors, while carbonization and nanoparticle doping are favorable for increasing their electrical conductivity for electrochemical sensor fabrication. Also, example fabrications and applications of natural fiber-based chemical sensors for chemical and biomarker detection are discussed. The selectivity of the sensors can be introduced and improved by surface modification of natural fiber, such as enzyme immobilization and biorecognition element functionalization, illustrating the adaptability of natural fiber as a smart sensing device, e.g., wearable and portable sensors. Ultimately, the high performances of natural fiber-based chemical sensors indicate the potential uses of natural fiber as a renewable and eco-friendly substrate material in the field of chemical sensors and biosensors for clinical diagnosis and environmental monitoring.

Strategy, Design, and Fabrication of Electrochemical Biosensors: A Tutorial

Advanced healthcare requires novel technologies capable of real-time sensing to monitor acute and long-term health. The challenge relies on converting a real-time quantitative biological and chemical signal into a desired measurable output. Given the success in detecting glucose and the commercialization of glucometers, electrochemical biosensors continue to be a mainstay of academic and industrial research activities. Despite the wealth of literature on electrochemical biosensors, reports are often specific to a particular application (e.g., pathogens, cancer markers, glucose, etc.), and most fail to convey the underlying strategy and design, and if it is transferable to detection of a different analyte. Here we present a tutorial review for those entering this research area that summarizes the basic electrochemical techniques utilized as well as discusses the designs and optimization strategies employed to improve sensitivity and maximize signal output.

Optimisation of electrochemical sensors based on molecularly imprinted polymers: from OFAT to machine learning

Molecularly imprinted polymers (MIPs) rely on synthetic engineered materials able to selectively bind and intimately recognise a target molecule through its size and functionalities. The way in which MIPs interact with their targets, and the magnitude of this interaction, is closely linked to the chemical properties derived during the polymerisation stages, which tailor them to their specific target. Hence, MIPs are in-deep studied in terms of their sensitivity and cross-reactivity, further being used for monitoring purposes of analytes in complex analytical samples. As MIPs are involved in sensor development within different approaches, a systematic optimisation and rational data-driven sensing is fundamental to obtaining a best-performant MIP sensor. In addition, the closer integration of MIPs in sensor development requires that the inner properties of the materials in terms of sensitivity and selectivity are maintained in the presence of competitive molecules, which focus is currently opened. Identifying computational models capable of predicting and reporting the best-performant configuration of electrochemical sensors based on MIPs is of immense importance. The application of chemometrics using design of experiments (DoE) is nowadays increasingly adopted during optimisation problems, which largely reduce the number of experimental trials. These approaches, together with the emergent machine learning (ML) tool in sensor data processing, represent the future trend in design and management of point-of-care configurations based on MIP sensing. This review provides an overview on the recent application of chemometrics tools in optimisation problems during development and analytical assessment of electrochemical sensors based on MIP receptors. A comprehensive discussion is first presented to cover the recent advancements on response surface methodologies (RSM) in optimisation studies of MIPs design. Therefore, the recent advent of machine learning in sensor data processing will be focused on MIPs development and analytical detection in sensors.

A new cysteamine-copper chemically modified screen-printed gold electrode for glyphosate determination

A chemically modified screen-printed gold electrode has been prepared by covering the electrode surface with a cysteamine-copper self-assembled monolayer (SAM). The sensor was effective for the voltammetric sensing of glyphosate. The method exploits the interaction of glyphosate with copper ions complexed by cysteamine, which results in a decrease in the intensity of copper redox current. Cyclic voltammetry was employed as a measuring technique. When dealing with voltammograms with numerous peaks changing in shape and size, it is difficult to define which signal is the most significant for the analyte determination; in these cases, a helpful approach is chemometrics. In this work, PLS (Partial Least Square regression) has been applied to build models to correlate the signal with the glyphosate concentration in standard aqueous solutions and tap water samples (matrix-matched calibration). The method's figures of merits were evaluated, obtaining a limit of quantification of about 5 μM. The reliability of the proposed sensor was verified by analyzing tap water spiked with glyphosate; recoveries higher than 90 % were achieved.

Capacitive tendency concept alongside supervised machine-learning toward classifying electrochemical behavior of battery and pseudocapacitor materials

In recent decades, more than 100,000 scientific articles have been devoted to the development of electrode materials for supercapacitors and batteries. However, there is still intense debate surrounding the criteria for determining the electrochemical behavior involved in Faradaic reactions, as the issue is often complicated by the electrochemical signals produced by various electrode materials and their different physicochemical properties. The difficulty lies in the inability to determine which electrode type (battery vs. pseudocapacitor) these materials belong to via simple binary classification. To overcome this difficulty, we apply supervised machine learning for image classification to electrochemical shape analysis (over 5500 Cyclic Voltammetry curves and 2900 Galvanostatic Charge-Discharge curves), with the predicted confidence percentage reflecting the shape trend of the curve and thus defined as a manufacturer. It's called "capacitive tendency". This predictor not only transcends the limitations of human-based classification but also provides statistical trends regarding electrochemical behavior. Of note, and of particular importance to the electrochemical energy storage community, which publishes over a hundred articles per week, we have created an online tool to easily categorize their data.

Machine learning-powered electrochemical aptasensor for simultaneous monitoring of di(2-ethylhexyl) phthalate and bisphenol A in variable pH environments

Plastic waste is a pernicious environmental pollutant that threatens ecosystems and human health by releasing contaminants including di(2-ethylhexyl) phthalate (DEHP) and bisphenol A (BPA). Therefore, a machine-learning (ML)-powered electrochemical aptasensor was developed in this study for simultaneously detecting DEHP and BPA in river waters, particularly to minimize the electrochemical signal errors caused by varying pH levels. The aptasensor leverages a straightforward and effective surface modification strategy featuring gold nanoflowers to achieve low detection limits for DEHP and BPA (0.58 and 0.59 pg/mL, respectively), excellent specificity, and stability. The least-squares boosting (LSBoost) algorithm was introduced to reliably monitor the targets regardless of pH; it employs a layer that adjusts the number of multi-indexes and the parallel learning structure of an ensemble model to accurately predict concentrations by preventing overfitting and enhancing the learning effect. The ML-powered aptasensor successfully detected targets in 12 river sites with diverse pH values, exhibiting higher accuracy and reliability. To our knowledge, the platform proposed in this study is the first attempt to utilize ML for the simultaneous assessment of DEHP and BPA. This breakthrough allows for comprehensive investigations into the effects of contamination originating from diverse plastics by eliminating external interferent-caused influences.

Intelligent Discrete Deep Learning Based Classification Methodology in Chemometrics

In recent years, deep learning models have attracted much attention for classification purposes in chemometrics. The popularity of deep learning models in this field comes from their unique features like universal approximation capability with the desired accuracy. Deep learning classifiers use several intelligent processing layers to model mixed, complex, and nonlinear patterns in the underlying data sets, which is why the development of deep learning based models has never been stopped in the chemometrics literature. Despite the variety of deep learning classification models used in this field, they all use a continuous distance-based cost function in their learning processes. Although using a continuous cost function for learning deep classifiers is a common approach, it conflicts with the discrete nature of the classification problem. In fact, applying a continuous cost function for inherently discrete classification problems can reduce the performance of the classification. In this research, a novel discrete learning based classification approach is proposed and implemented on a deep feed-forward neural network as one of the most commonly used deep learning models to develop a different learning process for deep classification models. The basis of the proposed learning approach is maximizing a discrete matching function of the actual and fitted values instead of minimizing a continuous distance-based cost function. The proposed classification approach is evaluated on five benchmark data sets in the chemistry field. The empirical results indicated the superiority of the proposed discrete deep learning approach over its classic continuous form. The results of this study demonstrate the important effect of discrete learning processes on the performances of deep learning classification models. Therefore, the proposed methodology can be a powerful alternative to common classification approaches to analyze chemical data in the chemometrics field.

Machine learning toward high-performance electrochemical sensors

The so-coined fourth paradigm in science has reached the sensing area, with the use of machine learning (ML) toward data-driven improvements in sensitivity, reproducibility, and accuracy, along with the determination of multiple targets from a single measurement using multi-output regression models. Particularly, the use of supervised ML models trained on large data sets produced by electrical and electrochemical bio/sensors has emerged as an impacting trend in the literature by allowing accurate analyses even in the presence of usual issues such as electrode fouling, poor signal-to-noise ratio, chemical interferences, and matrix effects. In this trend article, apart from an outlook for the coming years, we present examples from the literature that demonstrate how helpful ML algorithms can be for dispensing the adoption of experimental methods to address the aforesaid interfering issues, ultimately contributing to translate testing technologies into on-site, practical, and daily applications.

Nano- and Microsensors for In Vivo Real-Time Electrochemical Analysis: Present and Future Perspectives

Electrochemical nano- and microsensors have been a useful tool for measuring different analytes because of their small size, sensitivity, and favorable electrochemical properties. Using such sensors, it is possible to study physiological mechanisms at the cellular, tissue, and organ levels and determine the state of health and diseases. In this review, we highlight recent advances in the application of electrochemical sensors for measuring neurotransmitters, oxygen, ascorbate, drugs, pH values, and other analytes in vivo. The evolution of electrochemical sensors is discussed, with a particular focus on the development of significant fabrication schemes. Finally, we highlight the extensive applications of electrochemical sensors in medicine and biological science.

Bridging the Gap between Single Nanoparticle Imaging and Global Electrochemical Response by Correlative Microscopy Assisted By Machine Vision

The nanostructuration of an electrochemical interface dictates its micro- and macroscopic behavior. It is generally highly complex and often evolves under operating conditions. Electrochemistry at these nanostructurations can be imaged both operando and/or ex situ at the single nanoobject or nanoparticle (NP) level by diverse optical, electron, and local probe microscopy techniques. However, they only probe a tiny random fraction of interfaces that are by essence highly heterogeneous. Given the above background, correlative multimicroscopy strategy coupled to electrochemistry in a droplet cell provides a unique solution to gain mechanistic insights in electrocatalysis. To do so, a general machine-vision methodology is depicted enabling the automated local identification of various physical and chemical descriptors of NPs (size, composition, activity) obtained from multiple complementary operando and ex situ microscopy imaging of the electrode. These multifarious microscopically probed descriptors for each and all individual NPs are used to reconstruct the global electrochemical response. Herein the methodology unveils the competing processes involved in the electrocatalysis of hydrogen evolution reaction at nickel based NPs, showing that Ni metal activity is comparable to that of platinum.

Point-of-care biochemical assays using electrochemical technologies: approaches, applications, and opportunities

Against the backdrop of hidden symptoms of diseases and limited medical resources of their investigation, in vitro diagnosis has become a popular mode of real-time healthcare monitoring. Electrochemical biosensors have considerable potential for use in wearable products since they can consistently monitor the physiological information of the patient. This review classifies and briefly compares commonly available electrochemical biosensors and the techniques of detection used. Following this, the authors focus on recent studies and applications of various types of sensors based on a variety of methods to detect common compounds and cancer biomarkers in humans. The primary gaps in research are discussed and strategies for improvement are proposed along the dimensions of hardware and software. The work here provides new guidelines for advanced research on and a wider scope of applications of electrochemical biosensors to in vitro diagnosis.

Working at the interfaces of data science and synthetic electrochemistry

Electrochemistry is quickly entering the mainstream of synthetic organic chemistry. The diversity of new transformations enabled by electrochemistry is to a large extent a consequence of the unique features and reaction parameters in electrochemical systems including redox mediators, applied potential, electrode material, and cell construction. While offering chemists new means to control reactivity and selectivity, these additional features also increase the dimensionalities of a reaction system and complicate its optimization. This challenge, however, has spawned increasing adoption of data science tools to aid reaction discovery as well as development of high-throughput screening platforms that facilitate the generation of high quality datasets. In this Perspective, we provide an overview of recent advances in data-science driven electrochemistry with an emphasis on the opportunities and challenges facing this growing subdiscipline.

Spiers Memorial Lecture. Next generation nanoelectrochemistry: the fundamental advances needed for applications

Nanoelectrochemistry, where electrochemical processes are controlled and investigated with nanoscale resolution, is gaining more and more attention because of the many potential applications in energy and sensing and the fact that there is much to learn about fundamental electrochemical processes when we explore them at the nanoscale. The development of instrumental methods that can explore the heterogeneity of electrochemistry occurring across an electrode surface, monitoring single molecules or many single nanoparticles on a surface simultaneously, have been pivotal in giving us new insights into nanoscale electrochemistry. Equally important has been the ability to synthesise or fabricate nanoscale entities with a high degree of control that allows us to develop nanoscale devices. Central to the latter has been the incredible advances in nanomaterial synthesis where electrode materials with atomic control over electrochemically active sites can be achieved. After introducing nanoelectrochemistry, this paper focuses on recent developments in two major application areas of nanoelectrochemistry; electrocatalysis and using single entities in sensing. Discussion of the developments in these two application fields highlights some of the advances in the fundamental understanding of nanoelectrochemical systems really driving these applications forward. Looking into our nanocrystal ball, this paper then highlights: the need to understand the impact of nanoconfinement on electrochemical processes, the need to measure many single entities, the need to develop more sophisticated ways of treating the potentially large data sets from measuring such many single entities, the need for more new methods for characterising nanoelectrochemical systems as they operate and the need for material synthesis to become more reproducible as well as possess more nanoscale control.