Psychreg News Team

Health & Medicine
2 MIN READ

AI Develops Cancer Treatment in Just 30 Days and Predicts Patients’ Life Expectancy by Reading Doctors’ Notes

Cite This
Psychreg News Team, (2023, January 10). AI Develops Cancer Treatment in Just 30 Days and Predicts Patients’ Life Expectancy by Reading Doctors’ Notes. Psychreg on Health & Medicine. https://www.psychreg.org/ai-develops-cancer-treatment-just-30-days-predicts-patients-life-expectancy-reading-doctors-notes/
Reading Time: 2 minutes

In just 30 days, artificial intelligence (AI) has developed a treatment for an aggressive form of cancer and demonstrated its ability to predict a patient’s survival rate using doctors’ notes. These breakthroughs were performed by separate systems, highlighting how the powerful technology’s applications extend beyond image and text generation.

The University of Toronto collaborated with Insilico Medicine to utilise an AI drug discovery platform called Pharma to develop a potential treatment for hepatocellular carcinoma (HCC). The findings were published in the journal Chemical Science

The newly discovered treatment pathway for hepatocellular carcinoma (HCC), a form of liver cancer, was found by the AI, which also designed a “novel hit molecule” that could bind to the target. The system, created by scientists from the University of British Columbia and BC Cancer, can predict the survival rate of patients and has been found to be 80% accurate.

The use of AI technology is becoming a new weapon in the fight against deadly diseases. With the ability to analyze vast amounts of data, uncover patterns and relationships, and predict the effects of treatments, AI is proving to be a valuable tool.

Insilico Medicine founder and CEO, Alex Zhavoronkov, stated that while the world was focused on advances in generative AI for art and language, their generative AI algorithms were able to design potent inhibitors for a target with an AlphaFold-derived structure.

The team used AlphaFold, an AI-powered protein structure database, to synthesize a potential drug to treat hepatocellular carcinoma (HCC), the most common type of primary liver cancer. They accomplished this feat in only 30 days from target selection and after synthesizing seven compounds.

In a second round of AI-powered compound generation, researchers discovered a more potent hit molecule. However, any potential drug would still need to undergo clinical trials.

Feng Ren, Chief Scientific Officer and co-CEO of Insilico Medicine commented that AlphaFold broke new scientific ground in predicting the structure of all proteins in the human body.

The Insilico Medicine team saw this as an incredible opportunity to take these structures and apply them to their end-to-end AI platform to generate novel therapeutics to tackle diseases with high unmet needs. The paper represents an important first step in that direction.

Using natural language processing (NLP), a branch of AI that can understand complex human language, the system for predicting life expectancy analyzed oncologist notes after a patient’s initial consultation visit. The model identified individual patient characteristics and predicted their survival rate at six months, 36 months, and 60 months with more than 80% accuracy.

John-Jose Nunez, a clinical research fellow with the UBC Mood Disorders Centre and BC Cancer, explained that the AI reads the consultation document like a human would, extracting details such as the patient’s age, cancer type, underlying health conditions, past substance use, and family history to paint a complete picture of patient outcomes.

While traditional cancer survival rates have been calculated retrospectively based on a few general factors such as cancer site and tissue type, this model can pick up on unique clues within a patient’s initial 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 sites in British Columbia.

Nunez noted: “Because the model is trained on BC data, that makes it a potentially powerful tool for predicting cancer survival in the province.

“The great thing about neural NLP models is that they are highly scalable, portable and don’t require structured data sets. We can quickly train these models using local data to improve performance in a new region.”


The articles we publish on Psychreg are here to educate and inform. They’re not meant to take the place of expert advice. So if you’re looking for professional help, don’t delay or ignore it because of what you’ve read here. Check our full disclaimer