Chinese scholars develop new algorithms that accurately predict the outcome of mental illness treatment!

Release date: 2018-07-17

Can machine learning help with the diagnosis and treatment of mental illness? Recently, Dr. Bo Cao, a Chinese scholar from the University of Alberta, and his team completed a new study that can use machine learning techniques to predict the outcome of schizophrenia. The study was published in the journal Molecular Psychiatry, a journal of the Nature Publishing Group.

Schizophrenia is a serious mental illness accompanied by delusions, hallucinations and cognitive impairments. According to statistics, about one in every 100 people will be affected by schizophrenia. Most people with schizophrenia develop symptoms at a young age and will fight the disease for decades.

The early diagnosis of schizophrenia and many other mental illnesses, as well as the introduction of personalized treatment strategies, are an ongoing challenge for clinicians. At present, the treatment of schizophrenia is still obtained through trial and error. If the drug does not work, the patient may have long-term related symptoms and side effects, and is more likely to miss the best time to control and treat the disease. Therefore, it is important to study treatment options for diseases such as schizophrenia.

Researchers used machine learning algorithms to analyze newly diagnosed and untreated schizophrenia patients, as well as magnetic resonance imaging (MRI) of healthy subjects. By measuring the connectivity of other areas of the upper temporal cortex of the brain, the algorithm was able to successfully identify patients with schizophrenia with an accuracy of 78%. At the same time, the algorithm is able to predict whether the patient will respond positively to the treatment of the specific antipsychotic drug risperidone with an accuracy of 82%.

“This research is only the first step we have taken. We hope that we will eventually find reliable biomarkers that can predict schizophrenia before symptoms appear,” said Alberta, one of the leaders of the study. Dr. Bo Cao, Assistant Professor of Psychiatry at the University, said: "At the same time, we also want to use machine learning technology to optimize the treatment of patients, rather than let this technology replace the doctor. In the future, with the help of machine learning, doctors can be the first time It is a huge leap to choose the best medicine or treatment for the patient at the time of the visit."

Dr. Cao hopes to extend the scope of this research to other mental illnesses such as major depression and bipolar disorder. Although the initial results of this algorithm in the diagnosis and treatment of schizophrenia are exciting, the study still needs further validation in a large number of samples, and needs to be improved to improve the accuracy of prediction. In this way, the research results can be truly transformed into practical aids in the clinical environment.

Source: WuXi PharmaTech AI