Fact-checked by Grok 2 weeks ago

Electronic tongue

An electronic tongue is a multisensor system designed to mimic the human gustatory system, comprising an array of non-specific, low-selectivity chemical sensors that detect and analyze the chemical composition of liquid samples through cross-reactivity and pattern recognition algorithms.^[1] Introduced in 1995, it serves as an objective analytical tool for taste evaluation, primarily in liquid media, by generating electrical signals from sensor interactions with analytes and processing them via multivariate data analysis to classify tastes or product qualities.^[2] Unlike human tasters, it provides rapid, reproducible results without subjective bias, making it valuable for quality control and research.^[3] The concept of electronic tongues builds on early work in biomimetic sensors, with foundational developments in the 1990s by researchers such as Iiyama, Toko, Vlasov, and Legin, who pioneered potentiometric sensor arrays inspired by neural computing principles dating back to 1943.^[1] Key components include a sensor array (typically 4 to 40 units made from materials like lipid/polymer membranes, chalcogenide glasses, or noble metals), an interface for signal acquisition, and software employing techniques such as principal component analysis or artificial neural networks for data interpretation.^[1] Sensor types encompass electrochemical variants (potentiometric, voltammetric, impedimetric), optical, and bio-inspired designs, often enhanced with nanomaterials like graphene or carbon nanotubes for improved sensitivity.^[3] Electronic tongues find broad applications across industries, including food and beverage analysis for detecting adulteration, freshness, or flavor profiles in products like wine, honey, and milk; pharmaceutical development for assessing drug taste-masking and formulation stability; and environmental monitoring for identifying pollutants such as heavy metals or pesticides in water.^[3] In sensory science, they correlate highly with human panels (e.g., correlation coefficients often exceeding 0.95 for attributes like astringency), enabling non-destructive, high-throughput screening while reducing reliance on human subjects.^[2] Advancements as of 2025 include integration of artificial intelligence and hybrid systems (e.g., combined with electronic noses) to expand capabilities in diagnostics and precision agriculture, as well as graphene-based sensors achieving near-human taste discrimination and applications in virtual reality flavor simulation.^[3]^[4]^[5]

History

Origins and Early Development

The conceptual foundations of electronic tongue technology emerged in the 1980s, rooted in biomimicry of human taste buds and advancements in electrochemical detection of ions and compounds. Researchers drew inspiration from the biological gustatory system, where taste receptor cells respond to multiple stimuli with global selectivity rather than high specificity, leading to early investigations into lipid membranes as transducers for taste substances. These efforts built on electrochemical principles to mimic the interaction between taste compounds and cell membranes, focusing on potentiometric measurements to quantify basic taste qualities.^[6] Parallel developments occurred in Japan, with contributions from Shigeru Iiyama and collaborators on lipid/polymer membrane sensors, and in Russia, where Yuri Vlasov and Andrei Legin initiated work around 1994 on potentiometric sensor arrays for liquid analysis, later popularizing the term "electronic tongue." The first prototype of an electronic tongue was developed by Kiyoshi Toko at Kyushu University in Japan, with a patent filed in 1989 for a multichannel taste sensor using lipid/polymer membrane electrodes. This system employed potentiometric sensors coated with lipid/polymer membranes to detect the five basic tastes—sour, sweet, bitter, salty, and umami—by generating electrical responses proportional to taste substance concentrations. The prototype featured an array of several electrodes, each with distinct membrane compositions to provide differential signals for pattern recognition of taste profiles.^[7]^[8] Early experiments in the early 1990s centered on these potentiometric sensors, which emulated taste receptor cells through changes in membrane potential induced by taste ions and molecules. Initial studies demonstrated the system's ability to discriminate simple solutions, such as varying concentrations of salt (NaCl) and sugar (sucrose), by analyzing response patterns from the sensor array. Key publications from 1992 to 1994 highlighted these capabilities, including a 1992 study mapping taste profiles of beer samples using the multichannel sensor to differentiate bitterness and other attributes, and a 1993 experiment successfully classifying amino acids based on their taste qualities like umami and bitterness. These works established the feasibility of electronic tongues for qualitative taste assessment through statistical pattern analysis.^[9]^[10] By the mid-1990s, the development of electronic tongues benefited from cross-pollination with electronic nose technologies, particularly in adopting sensor array designs and multivariate data processing for handling non-specific responses. This influence enhanced the robustness of taste discrimination in complex mixtures, paving the way for broader applications while maintaining the core biomimetic focus.^[7]

Key Milestones and Commercialization

In the 2000s, electronic tongues progressed significantly through the integration of multisensor arrays with chemometric software for pattern recognition, facilitating the shift from laboratory prototypes to commercial viability. A notable example is the Alpha MOS ASTREE system, developed and marketed around 2000, which employs seven potentiometric ChemFET sensors to evaluate liquid samples for taste attributes like saltiness, acidity, and umami.^[11] This decade also saw applications in EU-funded initiatives, including the GOOD FOOD project (2004–2007), which advanced multisensor arrays for real-time food safety assessment, such as detecting contaminants in beverages and dairy products. A pivotal milestone came in 2005 with the IUPAC technical report, which formalized the definition of an electronic tongue as a multisensor system comprising low-selectivity sensors coupled with advanced mathematical tools for signal acquisition, processing, and analysis, prioritizing global pattern recognition over individual sensor specificity.^[12]^[13] During the 2010s, innovations expanded sensor modalities to include voltammetric and impedimetric techniques, enhancing sensitivity to diverse analytes such as redox-active compounds in complex matrices.^[14] Commercialization was propelled by collaborations between academic institutions and industry partners, exemplified by research at the University of Texas into scalable sensor technologies in the late 2000s, as well as the Insent SA402B taste sensing system—derived from Toko's Japanese efforts—for precise pharmaceutical taste profiling and formulation optimization.^[15]^[16]^[17] By the mid-2010s, electronic tongues had transitioned from specialized research instruments to routine tools in quality control, evidenced by a surge in scientific output exceeding 100 publications annually by 2015, underscoring their adoption in industrial settings for consistent product evaluation.^[18]

Principles and Components

Sensor Technologies

Electronic tongues rely on diverse sensor technologies to mimic human taste perception by detecting ions, redox-active compounds, and other taste-related analytes in liquid samples through electrochemical and optical mechanisms. These sensors form the core of the system, generating signals based on interactions with tastants such as acids, salts, and bitter substances, enabling the differentiation of basic tastes like sour, salty, bitter, sweet, and umami. The choice of sensors draws from established electrochemical principles, with adaptations for cross-sensitivity to handle complex mixtures without high specificity to individual compounds.^[19] Potentiometric sensors, the most common in electronic tongues, utilize ion-selective electrodes (ISEs) featuring lipid or polymer membranes that respond to specific ions by measuring potential differences across the membrane. These electrodes are particularly effective for detecting cations like Na⁺ (associated with saltiness) and H⁺ (linked to sourness), where the potential change is governed by the Nernst equation:

E = E_0 + \frac{RT}{zF} \ln a

Here, E is the measured potential, E_0 is the standard potential, R is the gas constant, T is the absolute temperature, z is the charge number of the ion, F is the Faraday constant, and a is the ion activity.^[20] This non-specific cross-response allows arrays of such ISEs to profile overall taste profiles in solutions. Early developments in the 1990s established potentiometric arrays as foundational for electronic tongues, with lipid-based membranes enhancing selectivity for taste ions.^[21]^[22] Voltammetric sensors operate by applying a varying potential to working electrodes, typically gold or platinum, and measuring resulting current transients to characterize redox-active tastants. Cyclic voltammetry, a prevalent technique, produces current-voltage curves where oxidation or reduction peaks reveal the presence of bitter compounds like quinine through distinct peak currents and potentials. These sensors provide dynamic information on electron transfer kinetics, complementing static potentiometric measurements for broader taste discrimination. Seminal work in the early 2000s demonstrated their utility in electronic tongues for food analysis, highlighting advantages in sensitivity to organic tastants.^[23]^[24] Impedimetric sensors quantify taste by monitoring changes in electrical impedance at sensor-solution interfaces, often using interdigitated electrodes to detect variations in charge transfer resistance or capacitance induced by tastants. Conductometric detection, a low-frequency subset, measures solution conductivity shifts for umami-related glutamates or astringent polyphenols. Optical sensors, employing fluorometry or absorbance spectroscopy, capture taste-induced changes in light emission or transmission, such as fluorescence quenching for astringency in tea infusions or absorbance for umami compounds. These methods extend detection to non-electroactive tastants, with impedimetric arrays showing robustness in complex matrices since the 2010s.^[25]^[26] Hybrid approaches integrate bioinspired elements into traditional sensors to improve selectivity and sensitivity, such as enzyme-immobilized electrodes for specific tastant recognition or nanomaterial modifications like carbon nanotubes on electrode surfaces since the 2010s. As of 2025, soft and flexible hydrogel-based sensors, inspired by saliva-like chemiresistive materials, have been developed to detect pungency in foods by mimicking protein binding mechanisms.^[27] Carbon nanotube hybrids enhance electron transfer and surface area, enabling detection of low-concentration analytes mimicking biological taste receptors. These bioelectronic designs, often combining voltammetric or potentiometric bases with biomolecules, have advanced electronic tongues toward greater mimicry of human gustation.^[28]^[29] Sensor arrays in electronic tongues typically comprise 5-20 diverse sensors, selected for overlapping sensitivities to generate multidimensional response patterns that capture the collective taste profile of mixtures. This design principle, rooted in cross-reactive sensing, allows pattern-based identification rather than isolated analyte detection, with array configurations optimized for specific applications through empirical validation. Integration of these arrays into portable systems facilitates real-time analysis.^[30]^[31]

System Architecture

The system architecture of an electronic tongue integrates a multisensor array with supporting hardware and software to enable collective sensing of chemical profiles in liquid samples. The core components typically include a sensor array housed within a probe, consisting of multiple electrochemical sensors (such as potentiometric or voltammetric types) that generate electrical signals in response to analytes.^[32] These arrays are paired with reference electrodes, often Ag/AgCl types, to provide a stable potential reference during measurements.^[33] Additionally, a pretreatment chamber is incorporated for sample conditioning, involving processes like filtration or dilution to reduce matrix interferences and enhance sensor reliability.^[32] The data acquisition unit forms the hardware backbone, featuring analog-to-digital converters (ADCs) and multiplexers that simultaneously process signals from the sensor array, converting analog outputs into digital data for analysis.^[33] Sampling rates vary by system and sensor type; for example, 1 Hz (one sample per second) is used in potentiometric measurements to capture steady-state responses.^[34] This setup ensures synchronized handling of multi-channel inputs, minimizing crosstalk and enabling real-time data logging. Interface layers facilitate sample-sensor interaction, primarily through flow cells for controlled liquid delivery or immersion setups where the probe is directly submerged in the sample.^[32] These are often equipped with automated stirring or pumping mechanisms to promote uniform analyte distribution and reproducible contact, critical for consistent signal generation across sensors.^[33] The software backbone relies on embedded microcontrollers to manage signal preconditioning, including amplification to boost weak sensor outputs and noise filtering to eliminate artifacts from environmental factors.^[33] This preprocessing occurs prior to data transmission for higher-level pattern recognition, ensuring clean inputs for subsequent analysis stages. Portability trends in electronic tongue designs have evolved significantly, shifting from bulky benchtop systems prevalent in the 1990s—often requiring laboratory power and manual operation—to compact, battery-powered configurations by the 2020s that support field deployment.^[35] These modern iterations, such as simplified potentiometric arrays integrated with data loggers, maintain analytical performance while enhancing mobility for on-site applications like water quality monitoring.^[35]

Operation and Data Analysis

Measurement Process

The measurement process of an electronic tongue involves a standardized sequence of steps to ensure reliable capture of taste-related signals from liquid samples, typically using a multisensor array immersed in controlled conditions. Liquid samples, such as beverages or pharmaceutical solutions, are prepared by dilution in aqueous media to volumes of 10-100 mL, often with co-solvents like ethanol to enhance solubility while maintaining physiological relevance, and filtered if necessary to remove particulates.^[36]^[1] This preparation occurs in a contamination-free environment, such as a sealed chamber, to prevent external influences on sensor responses. Sensors within the array, consisting of lipid/polymer membranes or electrochemical probes as described in the system architecture, are then immersed in the prepared sample either sequentially or simultaneously. Exposure duration ranges from 30 to 120 seconds, during which transient and steady-state signals—such as potentiometric potentials or voltammetric currents—are recorded to capture the dynamic interaction between the sample and sensor surfaces.^[36]^[7] Multiple acquisitions (e.g., 3-10 measurements per sample) are often performed, discarding initial readings to stabilize the signal and focus on reproducible steady-state data.^[36] Prior to each measurement series, a calibration routine baselines the sensors using standard taste solutions, such as quinine hydrochloride for bitterness or caffeine citrate, to verify response within specified ranges (e.g., relative standard deviation <4%) and compensate for potential drift via reference electrode potentials aligned with the Nernst equation.^[36]^[1] Following sample exposure, an automated cleaning cycle rinses the sensors with deionized water, ethanol-based solutions (e.g., 30% ethanol with HCl or KCl), or reference buffers for 10-60 seconds to eliminate residue and prevent carryover between analyses, with longer washes (up to 330 seconds) for thorough decontamination.^[7]^[1] The process culminates in the generation of raw output as multidimensional datasets, comprising voltage or current profiles (e.g., relative values calculated as sample potential minus reference potential, in mV) that represent the collective sensor responses for subsequent pattern analysis.^[36]^[7] These profiles, often including aftertaste metrics from post-cleaning reference measurements, provide a comprehensive snapshot of the sample's taste attributes without further processing at this stage.^[1]

Pattern Recognition Methods

Pattern recognition methods transform the multidimensional sensor responses from electronic tongues into interpretable taste profiles, enabling discrimination, classification, and quantification of samples. These computational techniques handle the inherent complexity and noise in data from sensor arrays, often comprising potentiometric, voltammetric, or impedimetric signals, to mimic human taste perception through statistical and machine learning approaches. Unsupervised and supervised algorithms are predominantly applied, with preprocessing ensuring data quality for reliable analysis.^[37]^[32] Data preprocessing is fundamental to mitigate variability, noise, and baseline drifts in raw sensor signals. Normalization via z-score standardization subtracts the mean and divides by the standard deviation for each feature, yielding data with zero mean and unit variance to facilitate comparison across sensors and samples. Feature extraction from transient signals, such as extracting maximum response, slope, or integral values from dynamic curves, further reduces dimensionality while preserving taste-relevant information. These steps enhance subsequent pattern recognition by addressing sensor drift and environmental influences.^[38]^[39] Unsupervised methods like Principal Component Analysis (PCA) perform dimensionality reduction and clustering without labeled data, revealing inherent structures in e-tongue datasets. PCA achieves this by projecting data onto principal axes that capture maximum variance, via eigenvalue decomposition of the covariance matrix \Sigma:

\Sigma = U \Lambda U^T

Here, \Lambda is the diagonal matrix of eigenvalues representing variances along each principal component, and the columns of U are the corresponding eigenvectors serving as the projection directions. This technique visualizes sample groupings, such as distinguishing beverage types, by retaining the first few components that explain most variance (often >90%).^[40]^[37] Supervised methods leverage labeled training data for targeted predictions. Partial Least Squares (PLS) regression models quantitative taste attributes, such as bitterness or sweetness levels, by constructing latent variables that maximize covariance between sensor inputs and taste responses, outperforming simple linear models in handling collinear data from e-tongue arrays. Artificial Neural Networks (ANN), including multilayer perceptrons, excel in classifying complex mixtures through non-linear mappings, achieving high discrimination rates for multi-component samples like wines or teas. For example, ANN models have demonstrated up to 99.14% accuracy in quality grading tasks.^[32]^[41] Linear Discriminant Analysis (LDA) serves as another supervised technique for taste discrimination, optimizing feature projections to maximize between-class separation while minimizing within-class variance. Cross-validation, such as leave-one-out, assesses model robustness by iteratively training on all but one sample, with e-tongue implementations frequently exceeding 90% accuracy; notable examples include 98.83% classification rates for wastewater samples using LDA.^[41]^[39] Validation of these models relies on metrics that quantify performance in blind tests. Confusion matrices tabulate true positives, false positives, and errors across classes, providing precision, recall, and overall accuracy for multi-class taste identification. Receiver Operating Characteristic (ROC) curves plot true positive rate against false positive rate, with the area under the curve (AUC) indicating discriminative power, often approaching 1.0 in optimized e-tongue systems. These tools ensure generalizability beyond training data.^[42]^[41]

Applications

Food and Beverage Industry

In the food and beverage industry, electronic tongues (e-tongues) play a crucial role in quality control by enabling rapid detection of spoilage and ensuring product consistency. For instance, AI-enhanced e-tongues have been developed to assess dilution levels and freshness in fruit juices, identifying spoilage through chemical changes with over 80% accuracy across various liquid samples.^[43] A 2025 study utilizing graphene-based sensors further demonstrated the ability to detect early spoilage in juices by analyzing volatile compounds and pH shifts, allowing for non-invasive monitoring during storage.^[44] Similarly, in beer brewing, e-tongues based on potentiometric ion-selective electrodes have classified different commercial beer types and predicted alcoholic strength since the early 2010s, to maintain batch uniformity.^[45] E-tongues facilitate flavor profiling in winemaking by providing non-destructive measurements of key attributes like tannins, acidity, and ripeness. In grape analysis, voltammetric e-tongues assess phenolic content to determine optimal harvest times, correlating sensor responses with sensory panel evaluations for varietal classification.^[46] This approach minimizes destructive sampling, enabling real-time adjustments in production to enhance wine quality and reduce variability across vintages.^[47] For shelf-life monitoring, voltammetric e-tongues track oxidation processes in oils and dairy products, predicting expiration dates with high predictive accuracy. In dairy applications, these systems analyze fresh milk samples over storage periods, detecting microbial growth and lipid oxidation to forecast shelf life.^[48] For edible oils, e-tongues detect adulteration and assess polyphenol content non-invasively.^[49] Adulteration detection represents another key application, where e-tongues identify counterfeit or diluted beverages through deviations in taste patterns. Potentiometric e-tongues have quantified additives in soft drinks, distinguishing authentic formulations from diluted versions by analyzing ionic profiles and sweetness levels with semi-quantitative precision.^[50] In fruit juices, these devices detect water or sugar syrup adulteration by pattern recognition, offering a faster alternative to chromatographic methods for regulatory compliance.^[51] In precision agriculture, e-tongues integrated with electronic noses monitor water quality for irrigation and detect nutrient levels in soil extracts, supporting sustainable farming practices and crop health assessment.^[3] Integration of e-tongues into large-scale production lines supports batch uniformity by automating sensory assessments, significantly reducing reliance on costly human taste panels. In beverage manufacturing, portable e-tongue systems enable inline monitoring of flavor consistency across production runs, correlating outputs with traditional quality metrics to streamline operations and lower evaluation expenses.^[52] This adoption has been particularly impactful in breweries and juice processors, where real-time data fusion enhances traceability and scalability without compromising product standards.^[53]

Pharmaceutical and Medical Uses

Electronic tongues have been instrumental in pharmaceutical development since the early 2000s for taste masking in oral formulations, particularly to reduce bitterness in pediatric medications and improve patient compliance.^[54] These systems assess bitterness intensity by correlating sensor responses to standard references like quinine hydrochloride equivalents, enabling the evaluation of masking agents such as sweeteners, flavors, or coatings in drugs like antibiotics and antihistamines.^[55] For instance, electronic tongue analysis has demonstrated up to 80% reduction in perceived bitterness for sublingual tablets containing bitter active pharmaceutical ingredients through the addition of non-medicinal ingredients like aspartame and acesulfame potassium.^[56] In formulation optimization, electronic tongues predict drug release profiles by simulating dissolution in saliva mimics, providing rapid in vitro data that correlates strongly with human sensory panels. Correlation coefficients exceeding 0.9 (e.g., R² = 0.910) have been reported between electronic tongue outputs and human taste assessments for bitter compounds such as H1-receptor antagonists, allowing developers to refine excipient compositions for controlled release and palatability.^[36] This approach facilitates the design of oral dosage forms, such as orodispersible tablets, where taste-masking efficiency is quantified without extensive human testing.^[57] Beyond drug development, electronic tongues support medical diagnostics by analyzing biofluids like saliva and urine for disease markers, including alterations in taste profiles indicative of conditions such as cancer. In saliva samples, voltammetric electronic tongues combined with machine learning have discriminated oral cavity cancer patients from healthy individuals with high accuracy (up to 95% sensitivity), detecting subtle electrochemical changes linked to biomarkers.^[58] Similarly, for urine analysis, potentiometric and spectroscopic electronic tongues have achieved over 96% accuracy in identifying bladder and prostate cancers non-invasively, leveraging pattern recognition to profile volatile and ionic metabolites altered by disease.^[59] Electronic tongues contribute to regulatory compliance in pharmaceutical testing, particularly for generic drug equivalence, by providing objective data on taste profiles that align with human panel results, thereby reducing the reliance on costly and ethically challenging animal or human trials.^[60] Their validated methods support bioequivalence demonstrations under frameworks like those from the FDA, where sensor arrays confirm formulation similarity in bitterness and release behavior for orally disintegrating generics.^[61] A notable recent advancement is the 2024 Penn State University electronic tongue, which integrates ion-sensitive sensors with artificial intelligence for high-accuracy identification of liquid compositions, enabling rapid screening of medication consistency and potential adulteration in oral liquids.^[43] This AI-enhanced system achieves over 80% accuracy in distinguishing subtle differences in samples, offering a scalable tool for pharmaceutical quality control.^[62]

Environmental Monitoring

Electronic tongues have been deployed for water quality assessment in natural bodies such as rivers and lakes, utilizing multisensor arrays to detect heavy metals like cadmium (Cd(II)), copper (Cu(II)), lead (Pb(II)), and zinc (Zn(II)), as well as pesticides including carbofuran and dichlorvos, and organic pollutants such as phenolic compounds. These systems employ potentiometric, voltammetric, and amperometric sensors to generate toxicity profiles through pattern recognition, enabling the discrimination of contaminant mixtures at trace levels. For instance, a potentiometric electronic tongue with ion-selective electrodes achieved detection of heavy metals like Cd(II) at trace levels in river water samples.^[63] Similarly, an amperometric bioelectronic tongue incorporating enzymatic biosensors quantified pesticides in river water with high accuracy using artificial neural networks.^[64] A voltammetric bioelectronic tongue monitored phenolic pollutants in wastewater, demonstrating real-time tracking of degradation processes.^[65] Portable electronic tongues facilitate on-site pollution tracking in industrial effluents, allowing discrimination of pH shifts, ionic imbalances, and heavy metal concentrations without laboratory intervention. These devices, often equipped with wireless radio modems for data transmission up to 15 miles, monitor effluents in real-time, with sensitivities reaching 5.2 × 10⁻⁸ M for Cu²⁺ and Pb²⁺ in waste streams near industrial sites. Potentiometric arrays have quantified ammonium, sodium, and potassium ions in polluted rivers at concentrations from 7.3 × 10⁻⁶ M to 1.5 × 10⁻² M, achieving correlation coefficients exceeding 0.9.^[66] Such portability supports rapid field assessments, outperforming traditional methods in deployment speed while maintaining low power consumption around 500 mW. Bibliometric analyses reveal over 50 studies on electronic tongues for environmental aqueous samples since 2010, reflecting a surge in applications for water quality monitoring with sensitivities down to parts-per-billion (ppb) levels for various contaminants.^[67] This growth underscores the shift toward real-time, low-cost alternatives to conventional techniques, with key contributions from voltammetric and potentiometric systems in profiling complex matrices.^[68] Integration with Internet of Things (IoT) enables wireless electronic tongues for continuous data logging in remote areas, providing alerts for anomalies such as algal blooms via toxin detection like microcystins. Autonomous systems, such as those deployed on unmanned vessels, facilitate long-term ocean and river monitoring with IoT frameworks for data transmission and analysis.^[69] Recent examples from 2023 include electronic tongues using screen-printed electrodes to detect pesticides at 0.02 ppm in compliance with EU Water Framework Directive limits (e.g., 0.0001 mg/L per pesticide), offering over 90% accuracy and faster results than chromatography.^[70] These advancements support regulatory adherence under directives like 2006/118/EC, enhancing ecological surveillance.^[71]

Mimicry of Human Taste and Limitations

Relation to Biological Taste Perception

The human gustatory system detects five basic tastes—sweet, sour, salty, bitter, and umami—via specialized receptors located in taste buds clustered within papillae on the tongue and oral cavity. Sweet taste is mediated by the heterodimeric G-protein-coupled receptors T1R2 and T1R3, which bind sugars and artificial sweeteners to initiate signaling cascades involving phospholipase Cβ2 and transient receptor potential channel M5 (TRPM5). Umami perception occurs through T1R1/T1R3 receptors responsive to amino acids like glutamate, while sour taste is transduced by PKD2L1 (a polycystin channel) and OTOP1 proton channels that detect acidity via hydrogen ion influx. Salty taste relies on epithelial sodium channels (ENaC), and bitter taste is sensed by approximately 25 T2R family receptors tuned to diverse aversive compounds. These receptors convert chemical stimuli into electrical signals through ion channels and second messengers, generating patterned neural activity relayed to the brainstem and cortex for taste interpretation.^[72] Electronic tongues approximate this receptor diversity not through specific molecular bindings but via arrays of cross-selective, low-specificity sensors that produce collective, multidimensional response profiles akin to holistic "taste maps." Each sensor interacts broadly with multiple analytes, mimicking the overlapping sensitivities of taste bud ensembles rather than isolated receptor-ligand interactions, and the resulting patterns are analyzed using multivariate methods to classify or quantify tastes. This global selectivity enables e-tongues to emulate the integrative output of biological taste cells without replicating their biochemical specificity. As defined by the International Union of Pure and Applied Chemistry (IUPAC), an electronic tongue comprises a nonspecific sensor array coupled with pattern recognition or multivariate calibration to process cross-sensitive signals from complex solutions.^[73]^[74] Key differences arise in perceptual integration: human taste perception fuses gustatory signals with olfactory, tactile, and thermal inputs via multisensory brain regions, yielding a rich flavor experience shaped by context and experience, whereas e-tongues isolate chemical signatures from dissolved tastants for objective profiling. Despite these limitations, e-tongues often correlate strongly with human sensory judgments; for instance, in wine analysis, potentiometric e-tongue responses aligned closely with expert panel ratings for attributes like bitterness and astringency, demonstrating reliable mimicry in blind assessments. This artificial taste paradigm prioritizes global selectivity for broad analyte detection, in contrast to the evolutionary fine-tuning of human receptors for specific survival cues, such as energy-rich nutrients or toxins.^[75] E-tongue architectures are evolutionarily inspired by the spatial organization of taste buds into clustered papillae, where distributed receptor responses converge on neural pathways for centralized decoding, though electronic systems forgo the biological adaptability, such as receptor plasticity and learning-mediated refinement seen in humans.^[76]

Challenges and Future Directions

Electronic tongues face several technical challenges that limit their widespread adoption. Sensor drift, caused by environmental factors or material degradation, necessitates frequent recalibration to maintain accuracy, posing ongoing maintenance burdens.^[77]^[78] Limited capability for detecting volatile compounds, which is better addressed by electronic noses, can necessitate hybrid systems for comprehensive analysis in complex matrices.^[79] Additionally, high initial costs for commercial systems restrict accessibility, particularly for small-scale users. Standardization remains problematic due to variability in sensor designs and outputs across vendors, complicating comparability and regulatory approval.^[36] Recent advancements address some of these issues by enhancing portability and analytical capabilities. In 2025, hydrogel-based in-tape electronic tongues emerged as a low-cost, disposable solution for rapid beverage classification, enabling sequential analysis of liquids like juices and coffees with high accuracy through ion-selective hydrogel pores.^[80] AI-enhanced models integrated with electronic tongues have also progressed, providing insights into the AI's decision-making processes—often termed "inner thoughts"—to improve flavor discrimination and early spoilage detection in products such as milk and wine.^[81]^[82] These developments, including a 2024 study reported in 2025, achieve high accuracy (>95%) in taste differentiation for subtle variations.^[83] Further 2025 advancements include voltammetric e-tongues for shelf-life evaluation and AI integration for pharmaceutical taste masking.^[48]^[84] Looking ahead, future directions emphasize hybrid systems and ethical integration. Combining electronic tongues with electronic noses could create comprehensive sensory arrays for multimodal analysis in food and environmental monitoring.^[85] Biohybrid approaches incorporating living cells promise greater adaptability to dynamic samples, though they require overcoming biocompatibility challenges. In pharmaceuticals, electronic tongues support reducing animal testing by simulating taste profiles in drug formulation, aligning with ethical shifts toward non-invasive methods.^[73] However, AI-driven applications in environmental monitoring raise data privacy concerns, necessitating robust safeguards to protect sensitive ecological datasets.^[86]

References

[1]
Electronic tongue: An analytical gustatory tool - PMC
It is a valuable tool for assessment and prediction of the taste of pharmaceuticals and related products. It replaces human panels in routine analysis.
[2]
[PDF] Considerations of the use of the electronic tongue in sensory science
The electronic tongue (etongue) is a biomimetic, multisensory system used for analyzing liquid samples, developed as a rapid, unbiased, and inexpensive ...
[3]
Electronic Tongues and Noses: A General Overview - PMC
Electronic tongues and noses have a wide range of applications, including food and beverage quality control, environmental monitoring, and medical diagnosis [17] ...Missing: definition | Show results with:definition
[4]
Research and development of taste sensors as a novel analytical tool
Dr. Kiyoshi Toko is a University Professor at the Institute for Advanced Study, and a Professor at the R&D Center for Five-Sense Devices, Kyushu University. He ...
[5]
(PDF) Electronic Tongues–A Review - ResearchGate
This review examines the applications of electronic noses and tongues in food analysis. A brief history of the development of sensors is included and this is ...Missing: origins | Show results with:origins
[6]
JPH0354446A - Taste sensor and manufacture thereof - Google ...
PURPOSE:To provide a function similar to the taste of a human being by forming a film using the principal constituent of a taste receptive film said to be ...
[7]
Sensors and Materials
pp. 145-152. S&M106 Research Paper of Special Issue Published: 1992. Taste Map of Beer by a Multichannel Taste Sensor [PDF] Kiyoshi Toko, Toshimi Murata, ...
[8]
Discrimination of Taste of Amino Acids with a Multichannel Taste ...
Taste of amino acids was studied using a multichannel taste sensor with lipid membranes as the transducer of taste substances.
[9]
An Application of Serially Balanced Designs for the Study of Known ...
The α-ASTREE e-Tongue instrument uses seven sensors to characterize taste signals associated with a liquid sample. The instrument was used to study eight test ...<|separator|>
[10]
Electronic tongue based on an array of metallic potentiometric sensors
This work has been supported by GOOD FOOD Integrated project by EU. Authors would like to acknowledge the students of electronic engineering department of ...
[11]
Electronic Tongues Employing Electrochemical Sensors - 2010
Jul 9, 2010 · This review presents recent advances concerning work with electronic tongues employing electroanalytical sensors.
[12]
Artificial Tongue - Electrical & Computer Engineering at UT Austin
Jan 8, 2009 · Neikirk's tongue uses microspheres, tiny sensors that change color when exposed to a specific targets, such as certain kinds of sugars. The ...
[13]
About us - Insent | Intelligent Sensor Technology, Inc.
After production of PC-9800 computers stopped in the late 1990s, we developed the SA402B taste sensing system using IBM PC/AT-compatible computers. Several ...
[14]
Performance qualification of an electronic tongue based on ICH ...
Two electronic tongue systems are commercially available: the taste sensing system SA402B (Insent Inc., Atsugi-chi, Japan) and the α-ASTREE e-tongue (Alpha ...
[15]
Number of scientific publications containing “electronic tongues”...
Number of scientific publications containing “electronic tongues” (E-tongues) and “electronic noses” (E-noses) terms published per year. Information ...
[16]
Sensor Arrays and Electronic Tongue Systems - ResearchGate
Aug 7, 2025 · This paper describes recent work performed with electronic tongue systems utilizing electrochemical sensors. The electronic tongues concept ...
[17]
Potentiometric Electronic Tongues for Foodstuff and Biosample ...
Abstract. Potentiometric sensors are attractive tools for the fabrication of various electronic tongues that can be used in wide area of applications, ...Missing: seminal papers
[18]
The features of the electronic tongue in comparison with the ...
Sep 21, 1999 · The features of the electronic tongue in comparison with the characteristics of the discrete ion-selective sensors · 84 Citations · 9 References.<|separator|>
[19]
Voltammetric electronic tongues – basic principles and applications
This review discusses basic principles and applications of voltammetric electronic tongues. It is introduced by a description of the concept of electronic.
[20]
Fundamentals and application of voltammetric electronic tongues in ...
Electronic tongues (ETs) are bioinspired analytical tools based on the synergies between (bio)sensors and chemometrics.
[21]
Application of a microfluidic electronic tongue based on impedance ...
Here, we tested the efficacy of an impedimetric microfluidic e-tongue setup – comprised by four interdigitated electrodes (IDE) on a printed circuit board (PCB) ...
[22]
https://www.semanticscholar.org/paper/The-features-of-the-electronic-tongue-in-comparison-Legin-Rudnitskaya/b2c957e4088863289b1fd2c97f7f5a27a13f61eb
[23]
Nanoscale hybrid systems based on carbon nanotubes for ... - PMC
CNT-based biosensors functionalized with biomolecules, such as enzyme, antibody and olfactory receptor, can detect target molecules at a femto-molar ...
[24]
Bioelectronic Tongues Employing Electrochemical Biosensors
Aug 7, 2025 · This chapter presents recent advances concerning work with electronic tongues that employ electrochemical biosensors, that is, bioelectronic ...
[25]
Sensor Arrays and Electronic Tongue Systems - Wiley Online Library
Feb 8, 2012 · This paper describes recent work performed with electronic tongue systems utilizing electrochemical sensors. The electronic tongues concept ...
[26]
A priori tailored selection of sensor arrays for electronic tongues
Mar 1, 2023 · Herein we explore on the development of a simple methodology that allows the a priori selection of the optimal sensor array to carry out a specific application.
[27]
Applications of electronic nose (e-nose) and electronic tongue (e ...
This review summarizes the application of e-nose and e-tongue in determining the quality-related properties of foods.
[28]
Electronic Tongue—A Tool for All Tastes? - PMC - NIH
Dec 31, 2017 · The first electronic tongue that became available on the market was constructed by Toko and co-workers [6]. SA402B and TS-5000Z (Figure 2) ...Missing: Kiyoshi | Show results with:Kiyoshi
[29]
Analysis of Milk Using a Portable Potentiometric Electronic Tongue ...
Conclusion. In this work, a simplified and portable electronic tongue (PE-tongue) was developed and used to predict chemical characteristics of milk samples.
[30]
Critical View on the Qualification of Electronic Tongues Regarding ...
In this review, we aim to highlight the advantages, challenges, and limitations of electronic tongues (e-tongues) in pharmaceutical drug development.
[31]
A review on conjugated polymer-based electronic tongues
Aug 15, 2022 · Voltammetry has an advantageous signal-to-noise ratio because it is less susceptible to interference from electrical disturbances in comparison ...1. Electronic Tongue... · 2. Conjugated Polymers (cps) · 3. Conjugated Polymer-Based...
[32]
Comparison of multivariate preprocessing techniques as applied to ...
In an electronic tongue, preprocessing on raw data precedes pattern analysis and choice of the appropriate preprocessing technique is crucial for the ...Introduction · Electronic Tongue Setup · Data Preprocessing...
[33]
Signal Processing and Pattern Recognition in Electronic Tongues
Oct 25, 2025 · This chapter reviews the development of solutions related to the practical implementation of electronic tongue sensor arrays.
[34]
Principal Component Analysis of Transient Potential Signals from ...
This study investigates the potential of transient potentiometric signals generated by an array of ion-selective electrodes (ISEs) as the basis for a ...
[35]
A Systematic Review of the Applications of Electronic Nose and ...
This review explores the principles, advantages, and applications of e-nose and e-tongue systems in food quality assessment.
[36]
Application of Electronic Tongue for Detection and Classification of ...
The confusion matrices (Figure 4 and Figure 6) and performance metrics (Table 6 and Table 7) confirmed the high precision, specificity, and sensitivity achieved ...
[37]
A matter of taste: Electronic tongue reveals AI inner thoughts
Oct 9, 2024 · According to the researchers, the electronic tongue can be useful for food safety and production, as well as for medical diagnostics. The ...
[38]
Graphene-Based Electronic Tongue Identifies Spoiled Juice
Feb 12, 2025 · Electronic tongue sensors detect juice spoilage with AI, ensuring freshness and reducing waste in food and beverage production industries.
[39]
Beer classification by means of a potentiometric electronic tongue
Dec 1, 2013 · In this work, an electronic tongue (ET) system based on an array of potentiometric ion-selective electrodes (ISEs) for the discrimination of ...
[40]
Electronic Tongue Technology Applied to the Analysis of Grapes ...
This review explores the principles, development, and applications of ET and bioET in the wine industry, highlighting their capacity to assess grape ripeness.
[41]
Analysis of red wines using an electronic tongue and infrared ...
The cost of analysis, reduction in time, environment friendliness, and non-destructive nature of these techniques offers bright prospects for the future.Missing: ripeness blending
[42]
A new voltammetric electronic tongue method and its application in ...
Jan 15, 2025 · This study designed a “reference sample comparison method” for V-Et to assess the shelf life of fresh milk.
[43]
Voltammetric Electronic Tongues in Food Analysis - PMC - NIH
As already pointed out, voltammetric electronic tongues can provide valuable qualitative information about food samples when combined with a chemometric method ...
[44]
Semi-quantitative and quantitative analysis of soft drinks using an ...
Jun 20, 2011 · A potentiometric electronic tongue with 36 cross-sensibility lipo/polymeric membranes was built and applied for semi-quantitative and ...<|separator|>
[45]
(PDF) Quality Monitoring of Fruit Juices Using an Electronic Tongue
Aug 5, 2025 · It was also found that e-tongue was sensitive to adulterated juice samples (i.e., the addition of sugar syrup or water) as well as monitoring ...
[46]
Recent Applications of Potentiometric Electronic Tongue and ... - NIH
In this study, the detection of umami and astringency in chicken soup and soy sauce by the e-tongue and the trained panel was similar. However, it was found ...
[47]
Beer discrimination using a portable electronic tongue based on ...
A simple and portable electronic tongue (ET) system based on disposable SPEs was developed to create a tool capable of distinguishing between different types ...Missing: bock | Show results with:bock
[48]
Overcoming Challenges in Pediatric Formulation with a Patient ... - NIH
The taste masking in pediatric medicines for oral use is often crucial to improve their palatability. To this end, several taste masking strategies can be used, ...
[49]
Taste Sensor Assessment of Bitterness in Medicines - PMC - NIH
Jul 24, 2024 · This review article introduces its application in measuring the intensity of bitterness in medicine, confirming the achievement of taste masking.
[50]
Evaluation of the Masking Efficacy of Sweetening and/or Flavoring ...
The incorporation of a variety of sweetening and/or flavoring NMIs into a SL tablet of E could be shown to mask its bitter taste by up to 80%.
[51]
Taste masking analysis in pharmaceutical formulation development ...
In conclusion, the multichannel taste sensor or e-Tongue may be a useful tool to evaluate taste-masking efficiency for solution formulations and to compare ...
[52]
Using machine learning and an electronic tongue for discriminating ...
Jun 1, 2022 · In this study, we demonstrate that impedance data obtained with an e-tongue in saliva samples can be used to diagnose cancer in the mouth.
[53]
Analysis of urine using electronic tongue towards non-invasive ...
Jan 1, 2023 · Even though e-tongues are successfully commercialized, their application in cancer diagnosis from urine samples is underestimated. In this ...
[54]
Critical View on the Qualification of Electronic Tongues Regarding ...
Comparison of bitterness intensity between Prednisolone and Quinine in a human sensory test indicated individual differences in bitter-taste perception.
[55]
Electronic Tongue for Palatability of Dosage Forms
Electronic tongue analysis also means that taste evaluations can be incorporated into both stability studies and formulation development, potentially reducing ...Missing: correlation r> 0.9
[56]
Soda to milk: Electronic AI tongue detects food with 80% accuracy
Oct 10, 2024 · Researchers at Penn State have developed an extraordinary new tool: an electronic tongue capable of discerning subtle differences in liquids.Missing: medication consistency screening
[57]
https://www.sciencedirect.com/science/article/abs/pii/S0378517305008380
[58]
https://www.sciencedirect.com/science/article/abs/pii/S0039914022001230
[59]
https://www.sciencedirect.com/science/article/pii/S0956566322008508
[60]
(PDF) A Review of the Use of the Potentiometric Electronic Tongue ...
PDF | This paper introduces electronic tongue systems for remote environmental monitoring applications. This new approach in the chemical sensor field.
[61]
Overview of electronic tongue sensing in environmental aqueous matrices: potential for monitoring emerging organic contaminants
### Bibliometric Analysis and Summary of Electronic Tongue Studies
[62]
Electronic tongue applications for wastewater and soil analysis
May 20, 2022 · Mukherjee, et al. Rapid evaluation of integral quality and safety of surface and waste waters by a multisensor system (electronic tongue).Missing: tracking | Show results with:tracking
[63]
Internet of Things Enabled Electronic Tongue for Remote Monitoring ...
This study proposes a rapid wine provenance detection method based on the fusion information of electronic tongue (ET) and electronic nose (EN) combined with ..
[64]
https://doi.org/10.1016/j.bios.2007.10.013
[65]
Detection of Pesticides in Water through an Electronic Tongue and ...
Feb 5, 2023 · The objective of this study was to implement an electronic tongue with a set of screen-printed sensors for the detection and discrimination of ...2. Materials And Methods · 2.3. Data Processing · 3. Results And Discussion<|control11|><|separator|>
[66]
Taste and its receptors in human physiology: A comprehensive look
May 2, 2024 · This review comprehensively and systematically summarizes the current study about the sense of taste, the function of taste receptors, the taste–structure ...
[67]
Electronic Tongue - an overview | ScienceDirect Topics
15.2 Electronic tongue definition and principles. According to the IUPAC definition, an electronic tongue is “a multisensor system, which consists of a ...
[68]
NONSPECIFIC SENSOR ARRAYS (“ELECTRONIC TONGUE”) FOR ...
The first analytical device that resulted from these trends © 2005 IUPAC, Pure and Applied Chemistry 77, 1965–1983 Nonspecific sensor arrays 1967 Page 4 was an ...
[69]
The features of the electronic tongue in comparison with the ...
The aim of this paper is to compare the capabilities and sensor parameters of discrete sensors (ion-selective electrodes) with the data fitting by the ...Missing: seminal | Show results with:seminal
[70]
Tasting with an Electronic Tongue - AZoSensors
Dec 19, 2012 · These taste buds are separated at the throat and palate surface and clustered in the tongue as specific structures known as papillae.
[71]
Emerging trends of advanced sensor based instruments for meat ...
Dec 3, 2019 · Accurate signal transfer from e-tongue sensors is crucial for using the instrument but issues of signal drift have remained a challenge. Sensor ...Missing: recalibration | Show results with:recalibration
[72]
Electronic Nose and Tongue Technologies for Flavor and Quality ...
Oct 1, 2025 · This chapter attempts to discover the emerging area of biomimetic sensor know-hows precisely, Electronic Nose (E-nose) and Electronic Tongue (E- ...
[73]
A Systematic Review of the Use of Electronic Nose and Tongue ...
This review examines 112 studies published from 2004 to 2025, and examines the functionalities and performance in detecting various food product-associated ...Missing: projects | Show results with:projects
[74]
Hydrogel In-Tape Electronic Tongue - PMC - PubMed Central
Feb 26, 2025 · The Hydrogel In-Tape Electronic Tongue (HITS) is a sensor system using hydrogel tapes for rapid beverage classification, using a single pore ...
[75]
A matter of taste: Electronic tongue reveals AI 'inner thoughts'
Oct 9, 2024 · A recently developed electronic tongue is capable of identifying differences in similar liquids, such as milk with varying water content; diverse products.
[76]
A Sensitive Electronic Tongue Can Taste When Juice Starts to Go Bad
Jan 21, 2025 · A sensitive electronic tongue can taste when juice starts to go bad. An AI analysis and a chemical sensor determine drinks' dilution, freshness and type.
[77]
A matter of taste: Electronic tongue reveals AI inner thoughts
A recently developed electronic tongue is capable of identifying differences in similar liquids, such as milk with varying water content; diverse products, ...Missing: medication consistency screening
[78]
Applications of Electronic Nose, Electronic Eye ... - PubMed Central
Jan 6, 2023 · This review aims to provide an overview of the fundamentals of electronic nose (E-nose), eye (E-eye), and tongue (E-tongue), data preprocessing, chemometrics,Missing: compact | Show results with:compact
[79]
Exploring the ethical issues posed by AI and big data technologies ...
Oct 20, 2025 · The paper systematically analyzes the technological breakthroughs of AI and big data in drug R&D and deeply examines typical risk points such as ...