Through the use of machine learning, psychiatric researchers can apply analysis techniques to large amounts of data. This gives scientists the unprecedented opportunity to categorize and compare complex brain patterns, genes and behaviors to gain major insights into the nature of a specific mental illness.


A new Canadian study brings us closer to the idea that machine learning could one day play a major role in helping doctors diagnose and treat mental health disorders.


The study, published in the journalMolecular Psychiatry, used a machine-learning algorithm to look at functional magnetic resonance imaging (MRI) scans of newly diagnosedschizophreniapatients, previously untreated schizophrenia patients and healthy subjects.


By measuring the connections of the brain’s superior temporal cortex to other brain regions, the algorithm successfully identified schizophrenia patients with 78 percent accuracy. It also predicted with 82 percent accuracy whether or not a patient would respond positively to theantipsychoticdrug risperidone.


“This is the first step, but ultimately we hope to find reliable biomarkers that can predict schizophrenia before the symptoms show up,” said study leader Bo Cao, an assistant professor of psychiatry at the University of Alberta.

“这是第一步,但最终我们希望找到可靠的生物标记物,可以在出现症状之前预测精神分裂症。”阿尔伯塔大学精神病学助理教授,研究负责人Bo Cao如是说。

“We also want to use machine learning to optimize a patient’s treatment plan. It wouldn’t replace the doctor. In the future, with the help of machine learning, if the doctor can select the best medicine or procedure for a specific patient at the first visit, it would be a good step forward.”


Around one in 100 people will develop schizophrenia, a severe and disabling psychiatric disorder characterized by delusions, hallucinations and cognitive impairments. Most patients with schizophrenia develop the symptoms early in life and will struggle for decades.


According to Cao, early diagnosis of schizophrenia and many mental disorders is an ongoing challenge. Developing a personalized treatment strategy at a patient’s first visit is also a challenge for many clinicians.


Treatment is often determined by a trial-and-error style. If a drug is not working well, the patient may suffer prolonged symptoms and side effects, and miss the best time window to get the disease controlled and treated.


Cao hopes to expand the work to include other mental illness such as major depressiveand bipolardisorders. While the initial results of schizophreniadiagnosisand treatment are encouraging, Cao says that further validations on large samples will be necessary.


“It will be a joint effort of the patients, psychiatrists, neuroscientists, computer scientists and researchers in other disciplines to build better tools for precise mental health,” said Cao.


来源:University of Alberta Faculty of Medicine & Dentistry



  • 发表于:
  • 原文链接https://kuaibao.qq.com/s/20180802A127YF00?refer=cp_1026
  • 腾讯「云+社区」是腾讯内容开放平台帐号(企鹅号)传播渠道之一,根据《腾讯内容开放平台服务协议》转载发布内容。
  • 如有侵权,请联系 yunjia_community@tencent.com 删除。