AI may choose best depression therapy

A US research team has developed a computer that can accurately predict whether an antidepressant will work, based on patients’ brain activity.

The multi-site trial initiated by UT Southwestern in 2011 to better understand mood disorders — involving Stanford, Harvard and other institutions — demonstrates that artificial intelligence (AI) may soon help doctors objectively diagnose and prescribe depression treatments.

The researchers predict that tools such as AI, brain imaging and blood tests will revolutionise the field of psychiatry in the coming years.

“These studies have been a bigger success than anyone on our team could have imagined,” UT Southwestern psychiatrist Dr Madhukar Trivedi said.

“We provided abundant data to show we can move past the guessing game of choosing depression treatments and alter the mindset of how the disease should be diagnosed and treated.”

EEG-based predictions

Published in Nature Biotechnology, the study included more than 300 participants with depression who were randomly assigned a placebo or selective serotonin reuptake inhibitor (SSRI), the most common class of antidepressant.

An electroencephalogram (EEG) was used to measure electrical activity in the participants’ cortex before beginning treatment.

The team then developed a machine-learning algorithm to analyse and use the EEG data to predict which patients would benefit from the medication within two months.

In addition to the AI accurately predicting outcomes, further research suggested that patients unlikely to respond to an antidepressant were likely to improve with other interventions such as psychotherapy or brain stimulation.

The findings were validated in three additional patient groups.

“This study takes previous research, showing that we can predict who benefits from an antidepressant, and actually brings it to the point of practical utility,” Stanford University Psychiatry Professor Dr Amit Etkin said.

The team hopes to develop an AI interface that can be integrated with EEGs across the country and are seeking approval from the U.S. Food and Drug Administration.

Signatures of depression

Data from the study derive from the 16-week EMBARC trial, which Trivedi initiated at four sites to establish biology-based, objective strategies to remedy mood disorders.

The project evaluated patients with major depressive disorder through brain imaging and various DNA, blood and other tests. His goal was to address a troubling finding from another study he led (STAR*D) that found up to two-thirds of patients do not adequately respond to their first antidepressant.

“We went into this thinking, ‘Wouldn’t it be better to identify at the beginning of treatment which treatments would be best for which patients?’” Trivedi explained.

He said that EEG will likely be the most commonly used tool to examine brain activity because it’s less expensive than magnetic resonance imaging and — in most cases — will be equally or more effective.

Despite this, a blood test or MRI may be needed for some patients if the depression is manifesting itself in a different way.

“There are many signatures of depression in the body,” Trivedi said.

“Having all these tests available will improve the chances of choosing the right treatment the first time.”

Additional projects

While his team continues to evaluate data from the EMBARC trial, Trivedi has initiated other large research projects to help improve the remission rate of depression. Among them is D2K, a study that will enrol 2500 patients with depression and bipolar disorders and follow them for 20 years. In addition, RAD is a 10-year study of 2500 participants (aged 10–24 years) that will uncover factors for reducing the risk of developing mood or anxiety disorders.

Trivedi’s research team will assess patients’ biological signatures to determine the most effective treatment.

“It can be devastating for a patient when an antidepressant doesn’t work,” Trivedi said.

“Our research is showing that they no longer have to endure the painful process of trial and error.”

Image credit: ©stock.adobe.com/au/iaremenko