Fullscreen HTML5Med Res (??MB)HTML4

DIGITAL SCENT & TASTE electronic noses and tongues Image source: www.pexels.com/ photo/girl-sitting-on-grasssmelling-white-petaledflower-1879288/ By Dr David Maddison, VK3DSM Computers can do a lot of things that humans can now, but taste and smell are still firmly in our domain. Or are they? It may not be too long before your smartphone can alert you to odours, or you can see an image of a dish someone has cooked and then find out for yourself how it tastes. W e have five primary senses: hearing, sight, smell, touch and taste. Electronics can already interface readily with vision, hearing and touch, but what about the other two primary senses, smell and taste? Actually, electronics interfacing with those senses goes back further than you might think. But they have proven more difficult than the others. By the way, our other senses include balance, temperature, pain, time, hunger, thirst & proprioception, for a total of 10-20, depending on how you define them. Imagine watching an online video, a movie at a theatre or playing a computer game and experiencing the smell of a field of flowers or the smoke of a disaster. The taste and smell of food or spices could even be reproduced for a cooking show. We could also have an “electronic nose” that analyses smells for various reasons. Those would include digitising and synthesising those smells to reproduce them at another location, 12 Silicon Chip to check food for signs of degradation, or to ensure that batches of coffee or wine were consistent. Electronic noses could even be (and are) used for smelling patients to determine disease; dogs have been successfully trained to smell cancer from the unique chemicals that it produces. Parkinson’s disease is also said to produce a unique smell. Incidentally, the idea of using smell to detect disease is not new. The Ancient Greeks had people known as uroscopists who would smell and taste urine to determine disease conditions. The taste of urine was also used to detect diabetes until about the 1840s, when other tests were developed. In Australia’s Northern Territory, electronic noses are being investigated for detecting diseases in plants (see siliconchip.au/link/ac4k). Other possible or actual applications of electronic noses include: • ensuring batch consistency in food or other production processes Australia's electronics magazine • detecting fake or adulterated food and drink • checking the quality and monitoring the degradation of meat • checking raw food ingredients for freshness and contamination • checking the efficiency of cleaning processes • comparing different recipes or food manufacturing processes • comparing a food product with a competitor’s • determining the effect of substitution of one ingredient of a food product with another • detection of bacteria or other pathogens • detecting drugs or explosives • detecting land mines (as animals are used now) • finding truffles • detecting pollutants in the air or soil Some of these jobs (like checking food) are currently done by humans, but different people have different abilities in this field and some people siliconchip.com.au Fig.1: the location of the olfactory system. 1) Olfactory bulb. 2) Mitral cells. 3) Bone. 4) Nasal epithelium. 5) Glomerulus. 6) Olfactory receptor cells. Source: https://w.wiki/Cw9K Fig.2: the olfactory system in a typical vertebrate. Each olfactory receptor neuron (ORN) is attached to cilia; their odour receptors (ORs) are sensitive to one particular type of odourant. Source: www.frontiersin.org/systems_ neuroscience/10.3389/fnsys.2011.00084/full can’t experience certain tastes or smells. So having electronic devices to do these jobs would provide a great deal of consistency, among other benefits. In the future, an electronic nose could be made into a consumer product to check for contamination or adulteration of food and drink, especially when travelling in foreign countries with poor hygiene standards. Odour localisation is another possible application, which involves finding the source of a specific problem odour when it is not obvious. An electronic nose could potentially be used to map an area (say in a large building) to help locate the source of a bad smell. All the above comes under the auspices of “digital scent technology”. For sensing or producing taste, there is “gustatory technology”. Challenges Arguably, emulating a sense of smell or taste is more difficult than emulating vision or sound. Vision fundamentally involves sensing just one type of thing (photons), while a microphone involves detecting sound pressure waves. siliconchip.com.au In contrast, sense of taste or smell involves sensing hundreds or thousands of different types of molecules, and both smell and taste cannot easily be objectively defined. To synthesise or analyse smells and tastes, it is important to understand how our natural systems of smell and taste work. The olfactory system The system for sensing smells is known as the olfactory system. It is located in the nose, with smell perception being processed in the brain (see Fig.1). When we smell something, we are actually sensing chemical molecules, either of one type or a mixture. These chemicals cause the stimulation of dedicated nerve cells high up inside the nose called olfactory sensory neurons (OSNs) or olfactory receptor neurons (ORNs) – see Fig.2. Each neuron is connected to cilia (hair-like extensions), which have odour receptors (ORs) that are sensitive to a specific chemical. They behave like a lock and key. There are about 500 different types of odour receptors. Australia's electronics magazine ORNs connect to glomeruli, which connect to mitral cells. Mitral cells process information before conveying it to the brain, via electrical signals, where the smell is interpreted in the brain according to past experience. Odour sensation depends on the concentration of the chemicals that are sensed and their combination and type. As there are many different types of odour receptors, the sensation depends on the specific combination of chemicals sensed, unless it is a simple odour comprising a single type of molecule (eg, bleach). Gustatory system The gustatory system is responsible for the sense of taste, which is mainly perceived by specialised taste receptor cells of the taste buds on the tongue. There is a persistent myth that different areas of the tongue sense different tastes, but this was due to a misinterpretation of a 1901 paper by German scientist David P. Hänig and it has since been debunked (see https://w. wiki/Cs$d). May 2025  13 Today, we know that taste receptors are distributed all over the tongue, soft palate and even the throat; they are not confined to specific regions. While some parts of the tongue might be slightly more sensitive to certain tastes, the differences are negligible. The five basic tastes (sweet, sour, salty, bitter and umami) can be detected wherever there are taste buds. In addition to the tongue, taste perception is influenced by other senses such as smell (which is why things taste different or not at all if you have a blocked nose), texture, temperature of the food and even pain receptors incidentally activated with particularly spicy foods or with ‘cool’ tastes like menthol. Primary smells Just as there are primary colours from which all colours (red, green & blue) can be made, and there are primary taste sensations (sweet, sour, salty, bitter & umami), numerous primary smells there have been identified, from which many others can be synthesised (at least in theory). The concept of primary smells is not universally accepted and different classification schemes exist. According to one classification scheme (siliconchip.au/link/ac4d), the primary smells the human nose can detect are as follows: • Chemical: usually smells of synthetic origin such as ammonia, bleach, gasoline, paint etc. • Fragrant: eg, floral smells or certain spices. • Fruit: eg, banana, lime and orange (lemon is a ‘fresh’ smell often used in cleaning products). • Minty: eg, eucalyptus, camphor, mint and peppermint. • Pungent: eg, blue cheese, sweat, onions, garlic, some fermented products. • Sickening and decaying: eg, rotting flesh, sewerage, burning rubber, mercaptans (the odourant in natural gas and butane). • Sweet: eg, chocolate, caramel and vanilla. • Toasted/nutty: eg, almonds, peanut butter and popcorn. • Woody and resinous: eg, timber and natural resin smells. According to another classification scheme (https://w.wiki/7AMo), the primary smells are: • Musky: eg, perfumes. • Putrid: eg, rotten eggs. • Pungent: eg, vinegar. • Camphoraceous: eg, mothballs. • Ethereal: eg, dry cleaning fluid. • Floral: eg, roses. • Pepperminty: eg, mint gum. Odour intensity There is a suggested scale of odour intensity: 0 – no odour 1 – very weak (detection threshold) 2 – weak 3 – distinct 4 – strong 5 – very strong 6 – intolerable Advanced smell classification There are more complex smell classification schemes, such as the Leffingwell Odor Dataset, which contains the Fig.3: a Principal Odour Map, analogous to a colour map but much more complicated. Source: https://research.google/blog/digitizing-smell-usingmolecular-maps-to-understand-odor/ 14 Silicon Chip Australia's electronics magazine smells of 3423 molecules, described by experts. These were combined with another data set, GoodScents, to create the SMILES (Simplified Molecular Input Line Entry System) odour data set, which includes the smells of 4983 molecules described using 138 descriptors (siliconchip.au/link/ac4e). Such data sets are used for research and the classification of different smells, as determined by large numbers of people (large numbers are needed because people perceive smells differently). Another way to classify smells is to generate a Principal Odour Map (POM). Such a map is analogous to a colour map showing hue and saturation, but it is vastly more complex because there are far more parameters describing smells than light wavelengths. A POM contains a vast database generated by people who rate various smells. A particular smell might be described statistically by many descriptors. With the use of a neural network, they can be reduced to two principal components representing by two axes on a graph, as shown in Fig.3. In that example, 400 different molecules were described using 55 different labels. Smells of individual molecules are depicted by the grey dots. These dots can be grouped together into similar types of smells. Based on the smells and mapping of known molecular structures, the Odour Map can be used to predict smells of unclassified and unsmelled molecular structures. Natural vs artificial smell recognition Fig.4 shows the analogies between natural and artificial smell recognition. In a human, first there are the odour receptors on the cilia, which connect to the glomeruli and then the mitral cells in the olfactory bulb. Mitral cells process information before conveying it to the brain, where the smell is interpreted. The equivalent processes in an electronic nose use a transducer as the receptor and a signal processor to convert the output of the transducer to useful information. This information is then processed by an algorithm and a neural network to interpret the smell, providing an identification. As it is very difficult to associate particular molecules with particular siliconchip.com.au Receptor an m Hu Mucous Cribiform plate Cilia Olfactory nerve Interaction E-n Olfactory bulb Signal generation os Vestibular cortex Volatile compound Sensor array: Transducer Somatosensory cortex Gustatory cortex (taste) Visual cortex Olfactory cortex Auditory cortex Signal processing Resistance (Ω) e Brain olfactory cortex Processed signal Input Identification Output Red wine Pattern recognition Fig.4: a comparison of human and electronic smell sensing processes. Source: Electronic noses and disease diagnostics – www.nature.com/articles/nrmicro823 smells, electronic noses need to be trained using machine learning and artificial intelligence (AI) to associate a particular smell or group of smells with the one that the operator is interested in detecting. In the rest of this article, we will look at the history of smell reproduction, electronic noses (for analysing smells), electronic tongues (for analysing tastes), and finally, taste reproduction. Smell reproduction in cinema To add extra sensations to movies, various attempts have been made to add a sense of smell as follows. Some are even in current use. 1868 The Alhambra Theatre of Variety in London sprayed scent into the audience during a live theatre performance. 1906 or 1908 At the Family Theatre in Forest City, Pennsylvania, the scent of rose oil was blown towards the audience using an electric fan during the display of a film, possibly about the Rose Parade in Pasadena, California. 1916 The Rivoli Theatre in New York was equipped with a system of vents to blow scents into the audience during the playing of the movie Story of the Flowers. 1929 During the showing of the film Lilac Time (https://youtu.be/ mmeXUJl_RMk), lilac perfume was poured into the ventilation system of the Fenway Theatre in Boston towards the beginning of the film. Also in that year, during the showing of The Broadway Melody (https://youtu.be/ siliconchip.com.au oYSOl0qYVE0), a theatre in New York sprayed perfume from the ceiling. 1933 A system was installed to deliver odours during a screening at Paramount’s Rialto Theater in Broadway, New York. All the above attempts to introduce odours into films or plays were by manual means; the timing of the delivery was not integrated electronically into a film soundtrack or other automatic signalling system. One problem was that the smells could linger for a long time, sometimes days. The human nose also can’t quickly transition to the next smell until a previous one has cleared. This suggests an alternative, more personal delivery means would be ideal. Small amounts of an odourant could ideally be delivered close to a person’s nose and quickly cleared. This strategy was used in some future systems. 1939 Scentovision was developed by Swiss inventor Hans Laube and introduced at the New York World’s Fair. This was later to be renamed Smell-O-Vision. Up 32 different smells could be delivered at individual seats by a system of pipes, and the delivery timing and amount was controlled by the projectionist using a control board. The first film produced using this technology was Mein Traum. The odours delivered included hay, peaches, roses and tar, corresponding to on-screen action. After the one and only screening at the World’s Fair, the technology was Australia's electronics magazine seized by police on the pretext that a similar system was already licensed for use in the United States (www. imdb.com/title/tt0151530/trivia). Investors took the matter to court, but it was futile, and the investors lost their investment. It is not clear what this alternative system was. 1951 Emery Stern of New York was granted US patent 2562959 for an Fig.5: Emery Stern’s 1951 US patent (2,562,959) for a scent distribution system for motion pictures. A perforated film, running in synchronicity with the movie film, was to be used to select scents. May 2025  15 Fig.6 (left): a newspaper clipping from 1960 showing produce Michael Todd Jr and inventor Hans Laube with their SmellO-Vision device. Hans Laube is shown pointing to the vials which each contain a different scent. Those scents would be selectively projected through tubes to every seat in the theatre. Source: https://cinematreasures.org/photos/258071 Fig.7 (right): an illustration from US Patent 2,905,049 for Smell-O-Vision. The smell is contained in the cells (12), part of a ‘train’, which is advanced according to signals on the movie reel, detected by a light beam (45) and sensor (46). “electromechanical scent distribution to accompany a motion-picture”. He envisaged a system of scent containers (item 54 near the centre of Fig.5) that are selected by a system comprising a perforated reel running synchronously with the film reel. Information on when to release scents was encoded by holes, which would be detected photoelectrically to trigger scent release or stop it. Unfortunately, at the time, there was a craze for 3D films and wide screens, so this scene system was left by the wayside. 1953 General Electric announced Smell-O-Rama, but it was never used to make a film and the technology was not pursued. It was demonstrated with a 3D image of a rose and scented puffs from an atomiser. The lack of commercialisation may also relate to the craze for 3D films and wide-screen at the time. 1959 Smell-O-Vision (called Scentovision on its invention in 1939 by Hans Laube) was patented in this year – see Fig.6. About 30 different odours could be triggered by signals on the movie soundtrack. It was first used in 1960; it was expensive to install and was said to work erratically. Individual odours were placed in 16 Silicon Chip containers on a reel, which were connected into a ‘train’ that moved according to signals on the movie track past an air distribution system, to collect and distribute the odours. The train was wound onto a take-up reel (see Fig.7). Scents were delivered to individual seats. 1959 AromaRama was used by theatre pioneer Walter Reade Jr for the screening of Behind the Great Wall, which was not made with the use of AromaRama in mind. It was in colour wide screen with four-channel sound and 31-72 smells including earth, firecrackers, horses, incense, grass, oranges, restaurants, smoke and tea. The system used for AromaRama was similar to the 1951 patent by Emery Stern, but the scent track was contained on the film print itself and not a separate reel. In preparation for the next smell, the previous smell was neutralised by an electrostatic device called the Statronic, which removed the scent particles from the air (although the patent says a neutralising agent was used). Fig.8: an advertisement from 1960 for a movie featuring Smell-O-Vision. Source: www. filmaffinity. com/en/ film478082. html Australia's electronics magazine siliconchip.com.au Fig.9: the configuration of conducting polymer sensor arrays for electronic and bioelectronic nose sensors. Source: www.researchgate.net/figure/ fig3_51824845 Fig.10: an electrochemical gas sensor. Source: www.baseapp.com/ nodesense/wireless-gas-sensors Supposedly, the previous scent could be cleared within two seconds, but some observers disagreed. You can read an unfavourable 1959 review of the experience at siliconchip.au/ link/ac4f 1960 Smell-O-Vision was featured in the movie Scent of Mystery, the only movie ever made with this technology in mind – see Fig.8. It was released just weeks after Behind the Great Wall. The competition between the two was called “the battle of the smellies” by Variety magazine. 2006 Japanese communications company NTT, in co-operation with a Japanese film distributor, released smells during the showing of The New World. They were released at three rows of theatre seats designated “Premium Aroma Seats”. The aromas were contained within plastic balls that were mixed and released at appropriate times during the showing, as commanded by a controller connected to a computer. 2009 4DX is a multi-sensory theatre special effects system that produces various sensations delivered to the individual viewer. These include rotating and vibrating seats, a ‘leg tickler’, airflow, hot air and water spray onto the viewer, plus scents. Theatre-­wide, special effects such as fog, flashes of light, snow, wind can also be produced. There are several 4DX cinemas in Australia: • Village Cinemas – siliconchip. au/link/ac4l • Monopoly Dreams – siliconchip. au/link/ac4m • Event Cinemas – siliconchip.au/ link/ac4n Intrinsically conducting polymers are used, typically polyaniline, polypyrrole, or polythiophene. They can pick up gas concentrations greater than 10ppm and, unlike MOS sensors, do not require heating. These sensors are relatively easy to make and it is also relatively easy to vary the composition. They are probably the second most common devices used in eNoses after MOS sensors. For more details on conducting polymers, see our article on Organic Electronics in the November 2015 issue (siliconchip.au/Article/9392). Electrochemical sensors are small electrochemical cells, similar to a battery, but generally with three electrodes instead of two. The extra electrode is used for reference purposes. As a gas enters the cell, which contains a liquid or gel electrolyte, it changes the electrochemical characteristics of the cell, which can be measured as a change in potential – see Fig.10. They are not sensitive to all gases. Metal-oxide semiconductor (MOS) sensors contain a chemoresistive metal oxide coating, which changes its resistance in response to a target gas of interest (Fig.11). An array of MOS devices with different coatings may be used to make a device sensitive to a variety of odours. These are among the most popular sensor devices in electronic noses. Electronic noses may be purely electronic or bio-electronic. The purely electronic sensors respond to a variety of odour molecules, while in bio-electronic noses, an attempt is made to more closely mimic the operation of biological noses. Proteins are cloned from biological receptor molecules that bind to specific odour molecules. This high level of specificity allows for extremely high sensitivity. An important aspect of electronic noses is that they should be relatively inexpensive. The gold standard for measuring any gas mixture is gas chromatography mass spectrometry (GCMS), which is accurate and reproducible but expensive, and not amenable to make into a miniaturised portable device. Electronic noses use much simpler and cheaper technology by comparison. They may not be as good as GC-MS for identifying substances, but they are suitable for the purposes for which they are intended. A variety of different types of sensors have been used or proposed. They include: Conducting polymer devices are chemoresistive, which means they change their resistance in response to a gas of interest. They are specially formulated to respond to particular gases. An array of different polymers or compositions may be used to make a device sensitive to a variety of odours (Fig.9). Electronic noses (eNoses) An electronic nose can detect smells (and according to some definitions, flavours). The basic elements of an electronic nose are an odour collection system (equivalent to a nose in a human), odour receptors, signal processing and pattern recognition. siliconchip.com.au Fig.11: the working principle of MOS electronic nose sensors. Source: www.researchgate.net/figure/ fig1_361874229 Australia's electronics magazine May 2025  17 Fig.12: the Cyranose 320 electronic nose. Source: www.sensigent.com/ cyranose-320.html Fig.13: the Sensigent MSEM 160 electronic nose. Source: www. sensigent.com/img/pdf/MSEM%20 160%20Datasheet.pdf 18 Silicon Chip It is possible to have multiple MOS sensors on one die. The detection threshold of commercial versions of these types of sensors is around 1-1000ppm. A disadvantage is their high operating temperature of 150400°C, requiring onboard heating and resulting in relatively high power consumption. Nanocomposite arrays are composite materials in which two or more phases are present, at least one of the phases having dimensions in the nanometre (one millionth of a millimetre) range. The components are designed to adsorb odours of interest, causing a change in impedance that can be measured. One such device that has been produced uses the conducting polymer polyaniline in a nanostructured composite to detect ammonia in human breath; a sign of kidney disease. Optical sensors for electronic noses rely on the fact that different gases absorb different wavelengths of light. By passing a gas between an optical light source and receiver, and measuring the absorption at different wavelengths, the type of gas can be determined. Piezoelectric sensors or quartz crystal microbalance sensors use piezoelectric quartz crystals with coatings that adsorb molecules of interest. As they do so, the resonant frequency of the crystal changes, and that can be measured. An array of several such devices can be used, each sensitive to different gases, to analyse mixtures of gases. Photoionisation sensors are used to detect low concentrations of volatile organic compounds (VOCs). These sensors work by using UV light to ionise the gases of interest, creating positively and negatively charge ions. These ions result in a current flow, which can be measured. Surface acoustic wave (SAW) sensors are a type of device in which acoustic waves travel along the surface. A coating or nanostructured surface can be used that is sensitive to a particular odour. As it is adsorbed, the acoustic velocity changes and that can be measured. An array with a variety of coatings can be constructed so that different odours can be sensed. Commercial & experimental eNoses The Cyranose 320 is a handheld Australia's electronics magazine device from Sensigent (www.sensigent.­ com) that is designed to detect and identify complex chemical mixtures that constitute aromas, odours and fragrances (Fig.12). It can also be used to identify simple mixtures and individual chemical compounds. It uses an array of nanocomposite sensors as the sensing elements, which they call a “NoseChip”. Pattern recognition and training are used to teach the device to identify particular smells of interest to the user. According to the video at https:// youtu.be/r3jvpZPjcA4 the device can detect various pathogens and diseases in human breath. The MSEM 160 (Multi-Sensor Environmental Monitor) from Sensigent is a portable electronic nose that uses up to 30 different sensors, including nanocomposite sensors, electrochemical sensors, MOS sensors and photoionisation sensors (Fig.13). It is available with three different sensor configurations to detect: 1. Malodours like H2S (hydrogen sulfide), NH 3 (ammonia), CH 3SH (methyl mercaptan), organo-sulfur and organo-nitrogen compounds and mixtures. 2. Aromas like alcohols, aldehydes, terpenes and mixtures of volatile and semi-volatile organic compounds (VOCs). 3. Pollutants including CO, O3, NOx (nitrogen oxides), SOx (sulfur oxides) & other regulated gases and mixtures. It is also available in custom configurations. NTT Data (see https://nttdata-­ solutions.com/en/) is developing artificial nose technology controlled by artificial intelligence (AI). It is intended to determine questions such as should a public restroom facility be cleaned or not, what is the optimal expiry date for a food product, and quality control of coffee. The nose was entered in an SAP (a business analytics company) ‘hackathon’ competition and was tasked with smelling coffee samples. It used four sensors to measure various gas values, which became the unique ‘fingerprint’ of a smell. PEN3 is a portable electronic nose from Airsense Analytics (see https:// airsense.com/en) that uses ten different MOS sensors (Fig.14). Once trained for specific smells of interest, and with the use of its pattern matching algorithm, it is designed to give siliconchip.com.au Fig.14 (left): the PEN3 with the optional “enrichment and desorption unit (EDU)” under it. Source: https://airsense.com/en/ products/portable-electronic-nose Fig.15 (above): a smell.iX16 eNose chip. Source: https://smartnanotubes.com/products/ fast qualitative answers such as good/ bad or yes/no. Suggested uses are in process control, quality control and environmental monitoring. SmartNanotubes (see https:// smart-nanotubes.com) has developed a multi-channel electronic nose gas detector chip for the mass market. The chip, which is called the smell. iX16 (Fig.15) contains nanostructured materials that can detect multiple gases, smells and volatile organic compounds (VOCs). The chip uses just 1µW of power. These chips have been incorporated into a development kit with a device called the smell.Inspector, shown in Fig.16. AI-based software called smell. Annotator analyses detected odours from the smell.Inspector and provides information. The eNose Company (see www. enose-company.com) has developed an electronic nose specifically for detecting disease, shown in Fig.17. It uses a variety of sensors, including MOS, conducting polymer sensors and quartz crystal microbalance senors. The device has been certified to detect lung cancer, COVID-19 & colon cancer and is under investigation for the detection of tuberculosis, pulmonary embolism, colorectal cancer, Barrett’s oesophagus, thyroid carcinoma, multiple sclerosis and rheumatoid arthritis. You can watch a video on it at https://youtu.be/6KUwcWdUGpY In 2002, Australian scientists at the University of New South Wales were reported to have developed an electronic nose that can detect truffles, but we can find no further details or reference to this. Electronic tongues (e-tongues) IBM HyperTaste is an experimental system that uses both electrochemical and AI technology to taste and analyse fluids (Fig.18). Proposed examples of use include checking the authenticity of food and drink products, quality control of food and beverages (Fig.19) and monitoring water quality. It consists of sixteen conductive polymer electrochemical sensors. Signals from the sensors are sent to software in a mobile device like a smartphone, whereupon the raw data is uploaded to a cloud AI server, analysed and classified. Fig.16: the smell.Inspector development kit. It contains four iX16 chips, visible on the left. Source: https://smart-nanotubes.com/produkt/ smell-inspector-developer-kit/ Fig.17: an electronic nose for disease detection from The eNose Company. Source: www. enose-company.com/wp-content/ uploads/2022/10/1665128402161.jpg siliconchip.com.au Fig.18: IBM’s HyperTaste device ‘tasting’ liquid in a glass. In this case, it identified a certain authentic gin out of several fake alternatives. Source: IEEE Spectrum – siliconchip.au/link/ac4o Australia's electronics magazine May 2025  19 Fig.19: a classification of various fruit juices and wines by the HyperTaste. Source: https:// dataconomy. com/wp-content/ uploads/2022/06/ HyperTasteAI-based-etongue-analyzesthe-chemicalcomposition-ofliquids-3.jpg In tests, the device has been able to identify different types of bottled mineral water, identified fruit juices by fruit type, detected counterfeit alcoholic beverages, identified wines by brand and place of origin and determined the intensity of coffee. It has also been used on the autonomous ship Mayflower to sample seawater. Producing specific odours For research purposes, specific odours can be produced with an olfactometer. This is a device that produces particular odours at precise concentrations for subjects to smell. The purpose is usually scientific research, to test the ability to smell certain odours or to detect odours, or for market research to test new products. An example of a commercial olfactometer used for research is shown in Fig.20. It appears to be a Burghart Research Olfactometer OL023 (see siliconchip.au/link/ac4g). The smells are released through a plastic tube, and the response of a test subject’s brain can be measured in a functional MRI (fMRI) machine. Producing specific tastes The following devices can be used to synthesise tastes by electrical stimulation of the tongue or by the delivery of chemicals. Professor Yen Ching-Chiuan at KeioNUS CUTE Center, Smart Systems Institute of the National University of Singapore, has developed an experimental digital taste stimulator that stimulates tastes using both electrical and smell stimuli (Fig.21). “Electric salt” is a device developed by Professor Homei Miyashita with the purpose of enhancing the salty flavour of food. This is to allow Japanese people, who are said to consume too much salt, to reduce the intake of salt while maintaining the desired taste. The devices are in the form of a spoon and a bowl. A chopstick device has also been developed. The tongue is electrically stimulated with a waveform at 0.1-0.5mA with an undisclosed voltage and shape. The intensity of the current can be adjusted by the user. The devices are said to increase the perception of the saltiness of food by 1.5 times. A gustometer is a device used in scientific research to deliver to the tongue a predetermined concentration and volume of a substance for taste testing, over a specified period. It is named after the gustatory stimulus that arises from a chemical which activates the taste cells of the tongue, resulting in the perception of flavour. The liquid under study is delivered to the tongue via a plastic tube. The device is used for studies of taste perception in people and animals and functional MRI can be used to study the brain’s response to various taste stimuli. An example is shown in Fig.22 and a subject under test in an MRI machine can be viewed at www.wur.nl/en/ show/gustometer.htm Fig.20: a research olfactometer at Wageningen University & Research. Source: www.wur.nl/en/show/olfactometer.htm Fig.21: an experimental digital taste stimulator. Source: https://cutecenter. nus.edu.sg/projects/digital-flavor.html Australia's electronics magazine siliconchip.com.au 20 Silicon Chip The Norimaki Synthesiser was an experimental Japanese device, invented by Professor Homei Miyashita of Meiji University, that simulated tastes. A device was placed in contact with the tongue, which had agar gels containing the five basic tastes: sweet, umami, bitter, acidic and salty (see Fig.23). These tastes can be considered analogous to the primary colours of light. A voltage could be supplied to individual taste gels (see www.dailymail. co.uk/sciencetech/article-8359459/). With no voltage applied, a user would experience all five tastes. If a voltage is supplied to one or more individual tastes, the cations (positively charged atoms or molecules) move away from the tongue to the cathode side, so that taste is minimised. The intensity of the sensation depends on the voltage and current supplied via the control panel. The device is said to be able to simulate almost any taste, but not fragrances or spicy flavours. A Norimaki is a sushi roll wrapped in seaweed, which the device resembles. You can watch a video on this device at https://youtu. be/7HIm4LoAZxU NTT DOCOMO, a large Japanese telecommunications company, has developed a technology to share tastes online. A taste is first analysed and converted to 25 parameters defining the taste, then transmitted by digital means and recreated from a palette of the five basic tastes (sweet, sour, salty, bitter and umami) using 20 types of base liquid. A proprietary algorithm is used to take into account different individual’s taste perceptions. Taste the TV (TTTV) is a lickable TV screen that allows users to experience various tastes that are sprayed onto it from a carousel of ten canisters (Fig.24). A plastic film is rolled over the screen to allow new tastes to be experienced and also for hygienic reasons between users. It is proposed to be used for taste competitions, for the training of chefs and to experience the tastes presented in a movie. It was invented by Professor Homei Miyashita from Meiji University in Tokyo who also invented the Norimaki Synthesizer mentioned above. We are hoping that this idea will not be combined with a touchscreen! SC siliconchip.com.au Fig.22: a gustometer made using off-the-shelf modular pump system components: (1) Cetoni BASE 120 module with five low-pressure syringe pump modules, (2) clamp, (3) computer-controlled solenoid valves, (4a) syringe holders, (4b) syringe piston holders, (5) upright support structure, (6) highprecision glass syringes, (10) tubing connections, (11) ferrules for tubing. Source: https://edspace.american.edu/openbehavior/project/novel-gustometer/ Fig.23: the end of a cylinder which is placed against tongue. The colours (food dye) are just to distinguish the different gels. Source: www.dezeen. com/2020/05/28/norimaki-synthesizer-device-taste-technology Fig.24: Prof. Homei Miyashita’s TTTV device. There are ten spray canisters (right) to apply various taste chemicals to an LCD screen (left). A roll of plastic film advances between tastes or between different users. Australia's electronics magazine May 2025  21