This new bipolar blood test uses AI

mental health bipolar
Mental health can be diagnosed using a blood test and AI math. The findings of the mathematical model were capable of predicting at a level of precision of over 90% whether an individual suffers from bipolar disorder, and whether they respond to lithium treatment.

Are you or someone you love suffering from bipolar disease but aren’t sure? If playing with dirt isn’t helping your mental health maybe you need some more serious intervention. A simple blood test to help get the treatment and attention needed is now on the way. Researchers at Haifa University have developed a simple blood test to identify people coping with bipolar disorder (formerly known as manic depression) and can predict the efficacy of lithium, the drug given to patients with this disorder.

The study was published in Molecular Psychiatry: “For the first time, the findings of the study enable us to use a blood test to find out – within a short timeframe of a few days, and at a relatively low cost – whether a person is suffering from bipolar disorder. We can also predict the efficacy of lithium, the drug given to people coping with bipolar disorder, and to adjust the medication individually,” explains Dr. Shani Stern of the University of Haifa, the corresponding author of the study.

Bipolar disorder, better known by its former name manic depression, is a chronic psychological disorder characterized by recurrent, sharp changes in mood, oscillating between extreme happiness and sadness and depression. The prevalence of the condition among the adult population around the world is between one and three percent and the average age of onset is 19 years.

At present, the evaluation of the level of bipolar disorder is undertaken by a specialist physician in the field of psychiatry, and includes questions intended to determine whether or not the patient has bipolar disorder and what is the most effective treatment in their case. One of the commonest treatments is lithium, but only around one-third of patients respond to this drug.

Dr. Stern notes that due to the strong similarity between manic depression and other disorders, such as schizophrenia, there is a risk of misdiagnosis, at least in the early stages. Moreover, at present there is no way to know in advance whether or not lithium will help an individual patient.

The current study was undertaken by Dr. Stern and her research team in collaboration with Dalhousie University and the Salk Institute. The study sought to examine whether it is possible to use a blood test to identify a person suffering from bipolar disorder and to predict the efficacy of lithium treatment for that individual. The study examined cells from three different groups in the population: people who are not suffering from bipolar disorder; people who are suffering from bipolar disorder and who respond to lithium treatment; and people suffering from the disorder who do not respond to lithium.

In the first stage of the study, white blood cells isolated from all the participants were examined; a cell culture is produced that can be maintained for a long period by infecting the cells with the EBV virus, which causes mononucleosis. In the second stage the researchers extracted RNA from the cells in order to understand which genes are expressed in each population and to identify genes that are expressed differentially.

90% reliable test

The findings show that 80% of the differences in gene expression are related to the expression of immunoglobins, the chief components of the immune system. “The most significant finding is that in people suffering from bipolar disorder, a difference was found at the expression level of genes of the receptors to the antibodies; this may explain the high rate of comorbidities. There is a known correlation between various psychiatric disorders and ancillary morbidity,” Dr. Stern noted.

After obtaining the biological findings, the researchers used a computational model based on artificial intelligence comprising neural networks; the AI model was developed at Dr. Stern’s laboratory for the purpose of verifying the findings. The researchers also used cells from other laboratories that executed the same biological processes as in the current study in order validate that the computational model also works for cells sampled in various laboratories.

The findings of the mathematical model were capable of predicting at a level of precision of over 90% whether an individual suffers from bipolar disorder, and whether they respond to lithium treatment. “This method can enable people coping with psychiatric disorders to save many months of suffering by adjusting the right medication for them,” Dr. Stern concluded.

Karin Kloosterman
Karin Kloostermanhttp://www.greenprophet.com
Karin Kloosterman is an award-winning journalist, innovation strategist, and founder of Green Prophet, one of the Middle East’s pioneering sustainability platforms. She has ranked in the Top 10 of Verizon innovation competitions, participated in NASA-linked challenges, and spoken worldwide on climate, food security, and future resilience. With an IoT technology patent, features in Canada’s National Post, and leadership inside teams building next-generation agricultural and planetary systems — including Mars-farming concepts — Karin operates at the intersection of storytelling, science, and systems change. She doesn’t report on the future – she helps design it. Reach out directly to [email protected]

Read More

LEAVE A REPLY

Please enter your comment!
Please enter your name here

TRENDING

AI data centers are triggering panic, instead of cleantech opportunities

AI may unintentionally become the economic engine that finally modernizes America’s aging grid. California is experiencing a massive AI data center boom, ranking 3rd in the U.S. with 227 operating centers and 54 more in development as of April 2026, according to Stanford.

24 7 renewable energy: how solar, wind, batteries and AI SaaS replace fossil fuels

A new report from the International Renewable Energy Agency based in Abu Dhabi makes something clear that many in the industry already suspected. When solar and wind are paired with battery storage, they can deliver reliable, round the clock electricity at costs that compete with, and often beat, fossil fuels.

How does Neuralink work?

Current goals are to treat neurological disease like Parkinson’s and restore autonomy to people with severe physical limitations by controlling exoskeletons and prosthetics. There’s also huge potential to improve cognitive abilities such as memory, attention, and languages.

Astro uses AI to help procure land for renewable energy

For oil-rich, environmentally vigilant Gulf states, Astro isn’t just another startup story. It is a blueprint for accelerating an energy transition that is now existential, not optional.

Kia’s bootcamp trains car mechanics on EVs and the future 

During Bootcamp 1.0 in 2025, their pilot program, Kia trained 87 professional mechanics, and 34 have already secured jobs at local dealerships. Another 50 trainees are currently in training in Morocco.

How to quiet noise from construction in your office

Streets need to be resurfaced in New York but the humming and grinding noise is unsettling. Noise is environmental pollution. 

EarthX and a blueprint for sustainable investing

Trammell S. Crow, a Dallas-based businessman and father of four, is focusing his efforts on impact investing, and media that focuses on saving the planet through EarthX.

Mining Afghanistan’s Mineral Discoveries Similar to Avatar

Now that American forces in Afghanistan are commemorating the longest period of any war that America has been involved in, including the 1965-73 Vietnam War, the recent discoveries of large and extremely valuable mineral and metal deposits may finally bring to light a reason to continue the presence of US fighting forces in this war torn and backward country.

From Pilot Plant to Global Stage: How Aduro Clean Technologies’ 2026 Expansion Signals a Turning Point for Chemical Recycling Investors Like Yazan Al Homsi

The company's Next Generation Process (NGP) Pilot Plant in London, Ontario, has officially moved into initial operating campaigns, generating the kind of structured, repeatable data that separates laboratory promise from commercial viability.

Nobul’s Regan McGee on Shareholder Value: “Complacency Is the Silent Killer” 

Why the governance framework designed to protect shareholders so...

Should You Invest in the Private Market?

startustartup Unlike public stock exchanges, which offer daily trading, strict...

How to build a 100-year-company

Kongō Gumi is a Japanese construction company, purportedly founded in 578 A.D., making it the world's oldest documented company. What can we learn about building sustainable businesses from them?

How AI Helps SaaS Companies Reduce Repetitive Customer Support Work

SaaS products are designed for large numbers of users with different levels of experience, and also in renewable energy.

Popular Categories