基于脑影像和机器学习的抑郁型精神分裂症早期定性诊断预测模型展望

An overview of the study about early qualitative diagnosis of the schizophrenia comorbid with depression based on multi-modal brain imaging

抑郁症状可发生于精神分裂症病程的任何阶段,严重破坏精神分裂症的预后。精神分裂症中抑郁症状和单纯抑郁症的临床表现特征和疗效也存在很大差异。多种证据提示,可能存在抑郁型精神分裂症这一特殊的生物学亚型,此亚型可能存在其特异性脑神经机制,表现为多种脑影像学表征的差异;并且这种特异性机制与单纯精神分裂症和抑郁症的脑神经机制都不相同,也与抗精神病药联用抗抑郁药的疗效差异有关。鉴于抑郁症状对精神分裂症患者的极大危害性,美国和英国著名学者呼吁要充分重视,在今后几年内亟须从多个角度加强对精神分裂症患者抑郁症状的特异性神经机制及治疗靶点的研究,以提高早期疗效,降低抑郁症状对精神分裂症患者造成的危害。基于以上推论和既往的研究,笔者提出选取首发精神分裂症患者进行随访研究,建立基于基线脑表征的早期定性诊断分类器、疾病严重程度相关的客观评价指标体系和疗效预测模型,为这一亚型的早期精准诊治提供客观依据。

Depressive symptoms could occur at any stage in the course of schizophrenia,and seriously damages the prognosis of schizophrenia.There are great differences in symptoms between depressive symptoms in schizophrenia and the clinical symptoms of simple depression,and so are in the treatment effect.A variety of evidence suggests that there may be a special sub-type of schizophrenia,which could be named depression-schizophrenia.This sub-type can be reflected in the specific brain features,which can be characterized by brain imaging techniques.Based on the above reasons and our previous studies,we intend to enroll the primary episode schizophrenia patients in a follow-up study,using multi-modal brain imaging and pattern recognition technologies,comparing the differences of the brain figures between the schizophrenia patients with or without depression,to reveal the specific brain characters of the depression-schizophrenia.At last we airn to reveal the specific therapeutic targets,set up the predict index and effect model which could provide an objective basis for the early precise diagnosis and individual treatment of the depression-schizophrenia.