Big data aids in exploring mental illness

Special to Western NewsLed by Schulich School of Medicine and Dentistry professor Dr. Lena Palaniyappan, a new study uses artificial intelligence to determine if patients with major depression and schizophrenia will respond positively or negatively to a four-week brain stimulation program. The international team’s goal is to treat 60,000 qualified participants worldwide by 2022.

Dr. Lena Palaniyappan, along with an international team of researchers, looks to provide relief or millions around the globe by pairing brain stimulation with artificial intelligence and big data to uncover the role of genetics in successfully treating mental illness.

Preceding the 20th century, mental illness was thought of as dangerous – its subjects mad, possessed and shunned from society. As technology revolutionized research, scientists developed a deeper understanding of the biological factors related to mental illness and vastly changed the dialogue about mental health and those affected.

Despite significant progress, two obstacles stand in the way of providing individuals the treatment they need to cure or manage their condition. Palaniyappan, a Schulich School of Medicine & Dentistry professor and Robarts Research Institute scientist, notes the heavy burden mental illness bares on society due to its early age of onset and its relentless prevalence.

One in five Canadians are currently coping with a mental illness. Seventy per cent of mental health problems have their onset during childhood or adolescence. Age aside, once mental illness takes hold, its symptoms can last a lifetime.

Palaniyappan’s focus is on translational research – bridging the gap between bench and bedside. His team is using neuroimaging to aid clinical decision-making, translating patient concerns into research questions – and hopefully answers.

“Our research group answers questions that are directly relevant to our patients’ everyday struggles. We are not working on research that is just fascinating to us. We’re looking at things that are important to patients,” Palaniyappan said.

The state-of-the-art brain imaging facilities at Robarts and its close proximity to the London Health Sciences Centre provide patients with access to the only research facility of its kind in Canada. Thanks to a $2.1-million federal grant, Palaniyappan, along with an international team, is undertaking a unique research project.

Launched in May 2019, the study uses artificial intelligence to determine if patients with major depression and schizophrenia will respond positively or negatively to a four-week brain stimulation program. The international team’s goal is to treat 60,000 qualified participants worldwide by 2022.

Just as there are a variety of mental-health conditions, there are also a variety of treatments. Some disorders can be cured and others managed, restoring quality of life. Some mental illnesses require patients to engage in a trial-and-error process with potential medications.

Schizophrenia is among the disorders with no such cure. Each case is different, and can involve positive and negative symptoms, or one or the other. Negative mental health symptoms inhibit the individual – they take away one’s ability to show emotion, feel motivated and participate in social situations.

“If a patient has these symptoms, they are very disabling,” Palaniyappan said. “They are the reason people do not get their job back or attempt to return to school – they lose motivation and the interests they have in life.”

Palaniyappan hopes to dig further into the determinants of deep depression and schizophrenia, and know confidently who will respond well to brain stimulation long before the treatment begins. When participants respond well, the negative symptoms lessen greatly and they gain back the motivation they lost.

Patients and their families have grown frustrated with current proposed solutions that fail. By getting treatment right the first time, patients have more buy-in, resources are saved and society takes one step closer to lessening the burden of mental illness.

“Mental-health problems are opportunities to better the human experience,” Dr. Palaniyappan said.

Using a network of global partners, Palaniyappan and his team are moving away from the traditional approach of asking a specific research question of a small sample. Rather, by using big data and machine learning of gene characteristics to subgroup patients, he can better understand their treatment needs on a larger scale.

While training at Stanley Medical College, Palaniyappan grew fascinated by the emotional and mental journey his patients experienced while unwell.

Wanting to better understand the treatment they received, he participated in a clinical trial, in which the researcher was interested in dispensing antipsychotic medication to participants with and without a mental-health condition. The study aimed to discover and compare the experiences of all participants. A young Palaniyappan had no idea how deeply this would impact him.

After a long three days on a small dosage of Haloperidol, one of the most commonly used antipsychotic medications, he felt emotionless. Neither happy nor sad. Restless. Legs shaking. A zero-emotion state. The journal he kept revealed the tragic side effects of such a treatment.

“The experience just opened my eyes and I realized that this type of treatment was not sufficient,” Palaniyappan said. “We are using treatments that are only controlling symptoms, and producing a lot of side effects for patients.”

Unlike Haloperidol, the four-week brain stimulation program produces few, if any, side effects.

“Magnetic waves stimulate the brain for a brief period of time. The technology we use allows us to specify an area of the brain we want to stimulate, sending magnetic pulses to activate the area.”

The sessions are 20 minutes long and do not result in discomfort for the patient. Many read or watch television during the process. Participants must undergo treatment five days per week. Without administering the treatment, not even the most advanced technology or practiced physicians can determine if it will be effective.

With multitudes of data, the artificial intelligence will set out to identify patterns in the genes of each participant. By identifying specific genes participants have in common, the goal is to match the subgroups to treatment response.

If successful in this three-year study, Palaniyappan plans to replicate it with different external groups in hopes of finding the same result. The end goal is the recognition of brain stimulation as a clinical method.

“This is the first time we’re offering this treatment,” Palaniyappan said. “Families who are struggling can now come to us. Seeing the progress naturally increases my motivation to continue making a difference.”