Any research that requires make sensing of large data sets from a variety of sources can benefit from artificial intelligence (AI). Cancer research, hopefully, is one of those areas.
NVIDIA has partnered with the National Cancer Institute and the US Department of Energy (DOE) to develop the CANDLE (Cancer Distributed Learning Environment) an AI-based “common discovery platform” to help fight this horrific disease, which is now the leading cause of death worldwide.
This is all part of the Cancer Moonshot initiative – announced by President Barack Obama during his 2016 State of the Union Address and led by Vice President Joseph Biden – that aims to deliver a decade of advances in cancer prevention, diagnosis and treatment in just five years.
CANDLE will tackle three major issues when it comes to cancer treatment:
1. It will pore through genomic data to find the genetic signatures in cancer DNA and RNA that predict their response to treatments.
2. It will accelerate the simulation of protein interactions to see how they create the conditions for cancer.
3. It will automatically extract and study “millions” of patient records to understand how cancer spreads and reoccurs, and accelerate the simulation of protein interactions to see how they create the conditions for cancer.
“GPU deep learning has given us a new tool to tackle grand challenges that have, up to now, been too complex for even the most powerful supercomputers,” said Jen-Hsun Huang, founder and chief executive officer, NVIDIA. “Together with the Department of Energy and the National Cancer Institute, we are creating an AI supercomputing platform for cancer research. This ambitious collaboration is a giant leap in accelerating one of our nation’s greatest undertakings, the fight against cancer.”
NVIDIA joins a growing number of companies looking to use AI to fight cancer. IBM Watson, as detailed in a 60 Minutes report on AI, can read 25 million published medical papers in about week, suggesting new cancer trials that human doctors may be unaware of. Microsoft’s Project Hanover is developing a machine learning approach to personalize drug combinations for Acute Myeloid Leukemia (AML), where treatment hasn’t improved in the past three decades. Google’s DeepMind Health initiative is exploring whether machine learning can reduce the amount of time it takes to plan radiotherapy treatment for such cancers.
In 2015, MIT developed a computational model designed to automatically suggest cancer diagnoses by learning from thousands of data points from past pathology reports. According to the researchers on this project, upwards of 5 to 15 percent of lymphoma cases are initially misdiagnosed or misclassified. This creates a significant problem since different lymphomas require dramatically different treatment plans.
Essentially, all these efforts hope to reduce the amount of time it takes to develop a diagnosis and treatment plan so more time can be spent on the actual treatment.
“Today cancer surveillance relies on manual analysis of clinical reports to extract important biomarkers of cancer progression and outcomes,” says Georgia Tourassi, Director of the Health Data Sciences Institute at Oak Ridge National Laboratory, which is one of several labs that will be working with NVIDIA. “By applying high performance computing and AI on scalable solutions like NVIDIA’s DGX-1, we can automate and more readily extract important clinical information, greatly improving our population cancer health understanding.”