Scientists from China have developed an algorithm that is based on the activity striped body diagnose schizophrenia with an accuracy of more than 80 percent. Also, this analysis allows to predict the susceptibility of patients to treatment with antipsychotic drugs. Anomalies in the functioning of the striped body correlate with the work dofaminergicheskoy system and the expression of genes that are associated with the risk of schizophrenia. Article published in the journal Nature Medicine.
Diagnosing schizophrenia is complicated by the very broad symptoms and the fact that the causes of this disease are poorly understood. The main hypothesis about the mechanism of schizophrenia is disruption of dopamine balance. Dopamine — a neurotransmitter with a wide range of functions, it is closely linked to the reward system in the brain. While schizophrenia is mostly prescribed antipsychotics, which block dopamine receptors and reduce the activity of this neurotransmitter.
Scientists believethat the structure, which plays a Central role in the development of schizophrenia is striatum — a group of basal nuclei of the cerebral hemispheres. Patients in this structure is often enhanced dopaminergic activity. Ang Lee (Ang Li) and his colleagues from the Chinese Academy of Sciences decided to use the anomalies of the functioning of the striped body as a marker for the diagnosis of schizophrenia. This did a fMRI of the brain of patients with schizophrenia 560 and 540 healthy people.
The researchers then trained a classifier based on support vector machine to determine the diagnosis (healthy or patients with schizophrenia) in three performance striped body: amplitude of low frequency oscillations, the activity of the links inside the striped body and external structures of the brain (more than 12 thousand items).
For each volunteer we calculated the ratio of the anomalous striped body position of the characteristics in the space of support vectors: the coefficient was negative and positive patients from control participants. In addition, the researchers conducted a cross-validation of the algorithm by testing it on data from different medical centers and different MRI systems. Also calculated, and the ratio of the anomalous striped body for patients with other mental illnesses.
The researchers tested whether rejection of the work of a striped body with increased activity of dopaminergic systems with the expression of genes that are associated with schizophrenia. This used the results of positron emission tomography and single photon emission computed tomography in healthy volunteers. Also analyzed data about the activity of the 43 genes that are associated with schizophrenia. The spatial expression of the markers dofaminergicheskoy system and selected genes compared with the distribution of the amplitudes of low frequency oscillations in the striatum.
The researchers found a number of differences in how the striped body between people with schizophrenia and healthy volunteers. For example, patients had increased amplitude of low frequency oscillations, as well as different (in comparison with the control group) ties striped body with other areas. An algorithm that is trained on this data, different from patients with schizophrenia with an accuracy of more than 80 percent. The indicator does not differ from the control groups for all investigated mental illness in addition to bipolar disorder — people with this diagnosis ratio was lower than in controls.