Research teams around the world have been working for years on enhancing the quality of the "electronic tongue" to create a version that closely mimics the human sense of taste. While the realization of this dream has encountered challenges, a South Korean research team claims to have successfully overcome them, reaching the closest electronic version that emulates the complexities of human taste perception.
The "electronic tongue" can differentiate between various tastes and quantitatively assess detailed characteristics, making it a potential tool for developing new products and monitoring quality. The system comprises:
-
Firstly, Sensor Devices: These are the fundamental components that simulate the taste buds found on the human tongue, designed to detect different taste compounds. The devices may vary based on the specific application, but they typically include various types of chemical or biological sensors sensitive to tastes like sweetness, saltiness, acidity, and bitterness. These sensors generate signals in response to the presence and concentration of certain taste compounds.
-
Secondly, Data Acquisition System: This part collects signals from the sensors, including techniques that convert sensor signals into digital data.
-
Thirdly, Data Analysis Tools: Once the data is collected, it is analyzed using sophisticated algorithms and machine learning techniques. This step interprets the sensor responses and compares them to known taste profiles, identifying the taste compounds in the sample.
Challenges on the Path to Achievement
With this concept of the electronic tongue, research teams worldwide have produced versions utilized in the food and beverage industry for quality monitoring, taste testing, and analyzing food and beverage components and flavors, as well as detecting contaminants and overseeing production processes.
It has also been employed to evaluate the taste and composition of medicines, which aids in developing formulations and ensuring consistent taste, along with detecting any discrepancies or impurities.
Some research is exploring its use in monitoring water quality, assessing environmental samples, and identifying and measuring various compounds within them, contributing to pollution monitoring efforts. Furthermore, there is exploration into its application in medical diagnostics, such as detecting biomarkers in body fluids or evaluating the taste of oral medications for patients with taste perception issues.
While "electronic tongues" have shown promising results in these fields, their widespread adoption remains limited due to several challenges, including:
-
Complexity of Taste Perception: Human taste perception is complex, involving interactions between taste compounds that can create synergistic or suppressive effects. Replicating this complexity accurately in an electronic system is challenging.
-
Reliability in Dynamic Conditions: Ensuring consistent and reliable results under changing or dynamic conditions presents a significant challenge, with external factors such as temperature changes, pH variations, or the presence of interfering substances potentially affecting taste detection accuracy.
-
Data Interpretation and Error Handling: Accurately and efficiently analyzing taste data to distinguish between tastes, especially subtle differences, and handling errors or inconsistencies in the data, requires advanced algorithms.
- Selectivity and Sensitivity: Achieving high selectivity and sensitivity in detecting certain taste compounds or patterns while avoiding interference from unrelated substances is another challenge. Discriminating between similar tastes or compounds can be difficult.
What Did the Koreans Do?
In an effort to improve the technology by overcoming these challenges, researchers from the Daegu Gyeongbuk Institute of Science and Technology and the Korea Advanced Institute of Science and Technology in South Korea have announced success in obtaining the closest electronic version to the human tongue. They publicized their findings in a study published in the journal ACS Applied Materials & Interfaces.
The secret to the breakthrough achieved by the Korean researchers lies in three features, unveiled in a press release by the website of the Daegu Gyeongbuk Institute of Science and Technology:
- Firstly: Integration of Sensory Devices and Deep Learning
The Korean research teams have created a system that effectively integrates taste sensors with deep learning technology. This integration allows for the simultaneous and precise measurement of saltiness, acidity, bitterness, and sweetness. Previous attempts focused more on the sensory aspect without deep integration with advanced machine learning or deep learning algorithms for taste analysis.
- Secondly: Enhanced Accuracy and Reliability
The Korean study addressed limitations in accuracy and reliability seen in previous studies by introducing a custom deep learning algorithm for taste analysis, significantly enhancing the technology's ability to classify different tastes with over 95% probability. Previous technologies lacked such advanced algorithms, leading to limitations in taste differentiation and evaluation.
- Thirdly: Application Potential
The Korean research highlights the system's broad applicability across various industries, including food, cosmetics, and pharmaceuticals—achievements not realized with previous versions, which did not exhibit such diversity in use across industries.
The research team tested their version with six different types of wine, conducting taste identification experiments. Their new version successfully classified the six different types of wine with an accuracy of over 95%, proving the wide applicability of this version. Its potential is expected to expand to include various fields such as food and beverage development, cosmetics, and pharmaceuticals.
Professor Kyung In Jang of the Robotics, Mechanical and Electronic Engineering Department at the Daegu Gyeongbuk Institute of Science and Technology stated, "We have confirmed that our new version can distinguish tastes with high probability through wine testing, and we will continue testing it in the food industry and various fields, such as cosmetics or pharmaceuticals."
Worthy Acknowledgement.. and Further Efforts
According to the results achieved in the Korean study, Mahmoud Mohamady, a nutrition professor at the Egyptian Agricultural Research Center, expresses admiration for the significant steps taken in overcoming the challenges facing electronic tongue technology. He notes that effectively combining taste sensors and deep learning technology to interpret different tastes is a commendable step towards emulating human taste.
Despite the deserving praise, he believes that asserting the latest version's proximity to human taste capabilities requires additional efforts from researchers for more extensive trials. Mohamady, in a phone conversation with Al Jazeera Net, states, "There's a need to test a larger number of compounds and expand the dataset for testing, conducting more comprehensive practical trials in the real world that go beyond the taste tests conducted by the researchers."
He also emphasizes that the potential applicability across various industries, including food, cosmetics, and pharmaceuticals, necessitates further studies and cannot be conclusively determined based on the Korean research trial alone.
He adds, "There may be aspects that have not been fully explored in terms of the ability to transfer this technology across diverse sectors, and this can only be discovered through practical trials."
