AI Could Cure Cancer

A new research document released in the US suggests with the help of artificial intelligence cancer could quickly become a disease of the past.

Scientists at the Pharmaceutical Artificial Intelligence (pharma.AI) Group, announced the publication of a paper demonstrating the application of generative adversarial autoencoders (AAEs) to generating new molecular fingerprints on demand.

Generative adversarial autoencoders are a relatively new branch of unsupervised machine learning, implemented by a system of two neural networks competing against each other in a zero-sum game framework.

In layman’s terms, the scientists have developed a way to target cancer in a more direct way than before.  

The study was published in US’ Oncotarget on December 22. It represents the proof of concept for applying Generative Adversarial Networks (GANs) to drug discovery, Eureka Alert reports.

The scientists now plan to launch a comprehensive drug discovery engine producing promising therapeutic treatments to significantly accelerate pharmaceutical R&D and improve the success rates in clinical trials.

Since 2010 deep learning systems demonstrated results in image, voice and text recognition, in many cases surpassing human accuracy and enabling autonomous driving, automated creation of art and even composition of music.

In recent years GANs produced extraordinary results in generating meaningful images according to the desired descriptions.

The new paper represents a proof of concept of an artificially-intelligent drug discovery engine, where computers are used to produce new molecular fingerprints with the desired molecular properties.

Earlier this year the pharmaceutical artificial intelligence division at the same company published several proofs of concept papers demonstrating the applications of deep learning to drug discovery, biomarker development, and aging research.

“Generative AAE is a radically new way to discover drugs according to the required parameters, adding “we use this pipeline to uncover the prospective uses of molecules, where these types of data are available,” said Alex Aliper, president, European R&D at the Pharma.AI group of Insilico Medicine.