AI and Electric Tongue Predict Drug Bitterness

Ever wondered why convincing a child to take their medicine can sometimes feel like trying to negotiate a diplomatic treaty? Or why even adults can be remarkably resistant to the idea of swallowing a pill, all thanks to a less-than-pleasant aftertaste? A group of researchers from the UK utilized an ‘electric tongue’ coupled with an ‘artificial intelligence (AI) system’ to forecast the bitterness of pharmaceuticals.

Taste plays a pivotal role in medication adherence, and finding a palatable solution has been a longstanding challenge in the field of healthcare.

From picky eaters to individuals with heightened taste sensitivity, the battle against bitterness in medicine is real. But with the 'electric tongue' and the power of artificial intelligence it is soon going to be a possibility.

Ensuring that individuals consistently take their medications is crucial, and taste plays a significant role in drug development. This is particularly evident in cases involving children, where taste has been identified as the primary obstacle to compliance.

However, it's also a concern for adults, especially those on long-term medications like HIV treatment (1^✔ ✔Trusted Source
Electronic Tongue—A Tool for All Tastes?

Go to source).

Electronic Tongue and Artificial Intelligence Couple to Determine Bitterness of Medicines

The team from University College London (UCL) employed an e-tongue, a device equipped with taste-sensitive sensors, to assign bitterness scores to medicines. This allowed them to estimate the expected aversiveness of the planned clinical dosage.

The e-tongue measures the extent to which bitter molecules adhere to a plastic sensor, mimicking the human tongue, and compares it with a clear sample. The difference between these measurements indicates the theoretical level of bitterness in a medicine. Using an e-tongue enables a faster and more efficient testing process compared to traditional human trials.

Advertisement

Additionally, the researchers collaborated with experts in machine learning to expedite drug development further using an AI model. By utilizing data from the e-tongue, the AI model deconstructs a drug into a set of molecular descriptors (such as the number of atoms, and total surface area of the molecule) that influence taste, ultimately predicting bitterness levels.

Is Making Medicines More Palatable in the Future Possible?

The forthcoming model will be an accessible resource, allowing pharmaceutical development globally to benefit from this data on medication palatability. Dr. Hend Abdelhakim from UCL Global Business School for Health explained that taste in medicines poses a particular challenge for children due to their heightened taste sensitivity.

This issue is amplified in cases of long-term illnesses like HIV, where antiretroviral medicines often have an unfavorable taste. Ensuring consistent intake of these pills is paramount, especially if treatment starts at a very young age. Dr. Abdelhakim emphasized that even if a drug is highly effective, its efficacy is compromised if the patient doesn't adhere to the prescribed regimen.

Advertisement

Maintaining adherence to antibiotic treatment is also crucial, as discontinuous courses contribute to the development of antimicrobial resistance. According to Dr. Abdelhakim, with antibiotics, non-compliance not only delays recovery but also exacerbates the global issue of antimicrobial resistance, affecting us all on a broader scale.

Reference:

1. Electronic Tongue—A Tool for All Tastes? - (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5872051/)

Source-Medindia