[ad_1]
Could the diagnosis and treatment of mental health disorders be helped one day through machine learning? New research from the University of Alberta brings us closer to this future through a study published in Molecular Psychiatry.
The research was conducted by Bo Cao of the Department of Psychiatry at the University of Alberta, with the collaboration of Xiang Yang Zhang at the University of Texas Health Sciences Center in Houston. They used an automatic learning algorithm to examine functional magnetic resonance imaging (MRI) images of newly diagnosed and untreated schizophrenia patients and healthy subjects. By measuring the connections of a region of the brain called the temporal cortex superior to other brain regions, the algorithm was able to identify patients with schizophrenia with an accuracy of 78%. He also predicted with 82% accuracy if a patient would respond positively to a specific antipsychotic treatment called risperidone.
"This is the first step, but we hope eventually to find reliable biomarkers that can predict schizophrenia before symptoms manifest themselves." We also want to use machine learning to optimize the plan. treatment of a patient.This would not replace the doctor.In the future, with the help of machine learning, if the doctor can choose the best medicine or procedure for a specific patient at the first visit, it would be a good step forward. "
Approximately one in 100 people will be affected by schizophrenia at some point in their lives, a serious and disabling psychiatric disorder that is accompanied by delusions , hallucinations and cognitive disorders. Most patients with schizophrenia develop symptoms early in life and struggle with them for decades.
According to Cao, early diagnosis of schizophrenia and many mental disorders is an ongoing challenge. Setting up a personalized treatment strategy during the first visit to a patient is also a challenge for clinicians. The current treatment of schizophrenia is still often determined by a style of testing and error. If a medication does not work properly, the patient may suffer from prolonged symptoms and side effects, and miss the best time slot for controlling and treating the disease.
Cao hopes to expand work to other mental illnesses such as major depressive disorders. and bipolar disorder. Although the first results of the diagnosis and treatment of schizophrenia are encouraging, Cao says that other validations on large samples will be needed and that further refinement will be required to improve accuracy before the work can be performed. translated into a useful tool in a clinical environment. It will be a joint effort of patients, psychiatrists, neuroscientists, computer scientists and researchers from other disciplines to build better tools for accurate mental health, "said Cao. "We have a group of computational psychiatry at the University of Alberta with a team of excellent clinicians and scientists to work collaboratively on this challenging problem."
This article has been re-posted to from documents provided by the University of Alberta. Note: Content may have changed for length and content. For more information, please contact the cited source.
Reference: Cao, B., Cho, RY, Chen, D., Xiu, M., Wang, L., Soares, JC, & Zhang, XY (2018). Prediction of treatment response and individualized identification of naive schizophrenia in the first episode using cerebral functional connectivity. Molecular Psychiatry, 1. https://doi.org/10.1038/s41380-018-0106-5
[ad_2]
Source link
