AI develops cancer drugs in 30 days and predicts survival

Artificial intelligence has developed a cure for an aggressive form cancer in just 30 days and demonstrated that it could predict patient survival using doctors’ records.

The breakthroughs were made by individual systems, but they show how the use of powerful technologies goes far beyond the creation of images and text.

Researchers at the University of Toronto have partnered with Insilico Medicine to develop a potential treatment for hepatocellular carcinoma (HCC) using AI drug discovery platform called Pharma.

HCC is a form of liver cancer, but artificial intelligence has discovered a previously unknown treatment route and developed a “new hit molecule” that could bind to this target.

The system, which can also predict survival, is the invention of scientists at the University of British Columbia and British Columbia who found the model to be 80 percent accurate.

AI has developed a cure for cancer in just 30 days from target selection and after synthesizing just seven compounds.

AI is emerging as a new weapon against deadly diseases as the technology is able to analyze vast amounts of data, identify patterns and relationships, and predict the effects of treatments.

This was stated by the founder and CEO of Insilico Medicine Alexey Zhavoronkov. statement: “While the world has been fascinated by advances in generative AI in art and language, our generative AI algorithms have succeeded in developing powerful target inhibitors with a structure derived from AlphaFold.”

The team used AlphaFold, an artificial intelligence (AI)-based protein structure database, to develop and synthesize a potential drug for the treatment of hepatocellular carcinoma (HCC), the most common type of primary liver cancer.

The feat was accomplished in just 30 days from target selection and after the synthesis of just seven compounds.

In a second round of AI-assisted compounding, researchers have found a more powerful hit molecule, though any potential drug still needs to pass clinical trials.

Feng Ren, Chief Scientist and Co-CEO of Insilico Medicine, said: “AlphaFold has opened new scientific frontiers in predicting the structure of all proteins in the human body.

“At Insilico Medicine, we saw this as an incredible opportunity to take these frameworks and apply them to our end-to-end AI platform to create new therapeutics for diseases with high unmet need. This article is an important first step in that direction.”

Another AI system identified characteristics unique to each patient, predicting survival at six, 36 and 60 months with over 80 percent accuracy.

The system used to predict life expectancy used natural language processing (NLP) — a branch of AI that understands complex human language — to analyze the oncologist’s records after the patient’s initial consultation visit.

The model identified characteristics unique to each patient, predicting survival at six, 36, and 60 months with over 80 percent accuracy.

John-Jose Nunez, a psychiatrist and researcher at the UBC Center for Mood Disorders and British Columbia Cancer, said in statement: “The AI ​​essentially reads the consultation paper the way a human would read it.

“These documents have a lot of details, such as the age of the patient, the type of cancer, underlying health conditions, past substance use, and family history.

“AI brings it all together to paint a complete picture of patient outcomes.”

Traditionally, cancer survival rates have been calculated retrospectively and classified by only a few general factors such as cancer location and tissue type.

The model, however, can capture unique clues in the patient’s original consultation document to provide a more nuanced assessment.

The AI ​​was trained and tested using data from 47,625 patients across all six BC cancer facilities located in BC.

“Because the model is trained on cancer data, this makes it a potentially powerful tool for predicting cancer survival in the province,” Nunez said.

‘[But] The advantage of neural NLP models is that they are highly scalable, portable, and do not require structured datasets. We can quickly train these models using local data to improve performance in a new region.”