Conditions such as Alzheimer’s, Parkinson’s, Huntington’s, and cystic fibrosis are suspected to be caused by misfolded proteins. Being able to predict a protein’s shape enables a greater understanding of its role within the body.

Previous techniques used for determining the shapes of proteins – such as cryo-electron microscopy, nuclear magnetic resonance, and X-ray crystallography – takes years and costs tens of thousands of dollars per structure.

AI, the researchers hope, will enable target shapes to be modelled from scratch without requiring previously solved proteins to be used as templates.

DeepMind calls their AI-powered folding efforts AlphaFold.

AlphaFold uses two different methods to construct predictions of protein structures:

1. 1. The first method repeatedly replaces pieces of a protein structure with new protein fragments, building on a technique commonly used in structural biology. A neural network invents new fragments.

1. The second method is called ‘gradient descent’ which is a mathematical technique applied to entire protein chains rather than pieces and makes small, incremental improvements.

Image Credit: DeepMind

DeepMind says its work is a successful demonstration of how AI can reduce the complexity of tasks such as protein folding; speeding up the diagnosis and treatment of some of the world’s most debilitating conditions.

In a contest organised by the Protein Structure Prediction Centre, AlphaMind was judged the winner among a total 98 algorithms by predicting the shapes of 25 out of 43 proteins. The runner-up, in comparison, could only predict three of the 43 proteins.

“For us, this is a really key moment,” said Demis Hassabis, co-founder and CEO of DeepMind. “This is a lighthouse project, our first major investment in terms of people and resources into a fundamental, very important, real-world scientific problem.”

 AI & >.

The post DeepMind is using AI for protein folding breakthroughs appeared first on AI News.

The AI learns from historical patient data to determine what the lowest doses and frequencies of medication delivered the desired results to shrink tumours.

In some cases, the monthly administration of doses was reduced to just twice per year while achieving the same goal. Based on a trial of fifty patients, treatments were reduced to between a quarter and half of the prior doses.

Pratik Shah, Principal Investigator at MIT Media Lab, says:

“We kept the goal, where we have to help patients by reducing tumour sizes but, at the same time, we want to make sure the quality of life — the dosing toxicity — doesn’t lead to overwhelming sickness and harmful side effects.”

Some of the side effects of cancer medication can do more harm than good to a patient’s quality of life. By implementing the AI’s treatment strategy, the least toxic doses can be used.

The current model focuses on glioblastoma treatment.

Glioblastoma is the most aggressive form of brain cancer, although it can also be found in the spinal cord. It’s more commonly found in older adults but can impact any age.

Sufferers are often given a life expectancy of up to five years. Doctors often administer the maximum safe dosages to shrink tumours as much as possible, but with side effects that can impact a patient’s quality of life over that period.

In a press release, MIT said:

“The researchers’ model, at each action, has the flexibility to find a dose that doesn’t necessarily solely maximize tumour reduction, but that strikes a perfect balance between maximum tumour reduction and low toxicity.”

“This technique has various medical and clinical trial applications, where actions for treating patients must be regulated to prevent harmful side effects.”

Reinforced learning was used for the model whereby the AI seeks ‘rewards’ and wants to avoid ‘penalties’ so it optimises all of its actions.

The model started by determining whether to administer or withhold a dose. If administered, whether a full dose or just a portion is necessary.

A second clinical model is pinged each time an action is taken in order to predict the effect on the tumour.

In order to prevent just giving frequent maximum dosages each time – the researchers’ AI received a penalty whenever it handed out full doses, or a medication too often.

Without the penalty in place, the results were very similar to a treatment regime created by humans. With the penalties, the frequency and potency of the doses were significantly reduced.

The full research paper can be found here (PDF)

What are your thoughts on using AI to improve cancer patients’ quality of life?

The post AI is helping to make treatment for cancer more bearable appeared first on AI News.

How long it takes for radiation therapy plans to be created today can take days. Individual maps need to be created for each patient to determine where tumours need to be targeted.

This lengthy process is frustrating for the patient, their loved ones, and medical professionals who’d love nothing more than to spend time saving lives instead of creating plans.

Engineering researcher Aaron Babier and his team have stepped-in with AI-based software to automate the process and cut down how long it takes for a radiation therapy plan to be created from days to hours, potentially even minutes.

The team – from the University of Toronto’s Department of Mechanical & Industrial Engineering – also includes Justin Boutilier, Professor Timothy Chan, and Professor Andrea McNiven. Each of the researchers sees radiation therapy design as an optimisation problem.

By analysing historical radiation therapy data, the AI behind the software applied it to an optimisation engine to develop treatment plans. When the plans their software tool created was compared with those manually created for 217 patients treated for throat cancer, they were almost indistinguishable.

The difference, however, is their AI-powered tool created the plans within 20 minutes.

Babier explained:

“Right now treatment planners have this big time sink. If we can intelligently burn this time sink, they’ll be able to focus on other aspects of treatment.

The idea of having automation and streamlining jobs will help make health-care costs more efficient. I think it’ll really help to ensure high-quality care.”

Most of us have some unwelcome connection to cancer. According to statistics, one in two people in the UK born after 1960 will be diagnosed with some form of cancer during their lifetime.

Babier has a personal vendetta against the disease. He shares when he was 12 years old his stepmom sadly passed away from a brain tumour.

“I think it’s something that’s always been at the back of my head. I know what I want to do, and that’s to improve cancer treatment,” he says. “I have a family connection to it.”

What are your thoughts on the use of AI in cancer treatment?

The post AI slashes cancer treatment plan creation to ‘mere hours’ appeared first on AI News.