An innovative artificial intelligence model has been created by a team of researchers at the CUNY Graduate Center that may greatly increase accuracy while speeding up and lowering the cost of the medication development process.

The new model, termed CODE-AE, can test novel pharmacological molecules to precisely predict efficacy in people, as explained in research, "A Context-aware Deconfounding Autoencoder for Robust Prediction of Personalized Clinical Drug Response From Cell Line Compound Screening," published in Nature Machine Intelligence.

In testing, it could also find potentially more effective tailored medications for over 9,000 patients. The method is anticipated to speed up drug development and precision medicine greatly.

Artificial Intelligence to Predict How Humans Responds to Drugs

To find safe and effective therapies and choose an existing medicine for a particular patient, accurate and reliable predictions of patient-specific reactions to a novel chemical molecule are essential. However, directly testing a drug's early effectiveness on humans is immoral and impossible.

Experts usually use cell or tissue models of the human body to assess a pharmacological molecule's therapeutic efficacy. Unfortunately, the treatment efficacy and toxicity in actual patients frequently do not match up with the pharmacological impact in a disease model.

This knowledge gap largely causes high prices and low rates of drug discovery productivity.

"Our new machine learning model can address the translational challenge from disease models to humans," Lei Xie, a professor of computer science, biology, and biochemistry at the CUNY Graduate Center and Hunter College and the paper's senior author, said per Phys.org.

Commonly Used Painkillers
(Photo : JACK GUEZ/AFP via Getty Images)
Commonly used painkillers, medicines based on Ibuprofen, an anti-inflammatory drug. According to a study, Ibuprofen would increase the risk of heart attack by almost a quarter.

ALSO READ: AI for Medical Imaging Made More Understandable Through the Use of Saliency Heat Maps; How Effective Is This Technique?

How CODE-AE Works

CODE-AE makes use of various current developments in machine learning and has a design that is biologically inspired. For instance, one of its elements generates images using Deepfake using similar methods.

You Wu, a Ph.D., student at the CUNY Graduate Center and a research co-author, claimed that the new model could address the issue of not having enough patient data to train a generic machine-learning model.

Wu stated that, although several techniques have been created to use cell-line screenings for forecasting clinical reactions, their results are inaccurate because of data inconsistency and contradictions.

CODE-AE, per SciTechDaily, could extract intrinsic biological signals masked by noise and confounding factors and effectively alleviate the data-discrepancy problem.

Therefore, CODE-AE greatly outperforms state-of-the-art techniques in predicting patient-specific medication responses only from cell-line chemical screens in terms of accuracy and robustness.

The next task for the research team is to establish a method for CODE-AE to accurately forecast the impact of a new drug's concentration and metabolization in human bodies. The researchers also pointed out that experts may modify the AI model to precisely anticipate adverse medication effects in humans.

RELATED ARTICLE: Scientists Developed a Search Engine-Like AI Algorithm That Can Identify Rare Diseases

Check out more news and information on Artificial Intelligence in Science Times.