Published papers

© Springer Nature Switzerland AG 2020. Recently, the number of people who are diagnosed as mental diseases is increasing. Efficient and objective diagnosis is important to start medical treatments in earlier stages. However, mental disease diagnosis is difficult to quantify criteria, because it is performed through conversations with patients, not by physical surveys. We aim to automate mental disease diagnosis in order to resolve these issues. We recorded conversations between psychologists and subjects to build our diagnosis speech corpus. Our subjects include healthy persons, people with mental diseases of depression, bipolar disorder, schizophrenia, anxiety and dementia. All of our subjects are diagnosed by doctors of psychiatry. Then we made accurate transcription manually, adding utterance time stamps, linguistic and non-linguistic annotations. Using our corpus, we performed feature analysis to find characteristics for each disease. We also tried automatic mental disease diagnosis by machine learning, while the number of sample data is few because we were still in our pilot study phase. We will increase the number of subjects in future.

Interest in the microbiota‐gut‐brain axis, in which the intestine and the brain interact with each other via the microbiome and its metabolites, is rapidly growing. Autism spectrum disorder (ASD) is one of the most reported microbiota‐related disorders in the neuropsychiatric field. In ASD, differences in specific bacterial species and/or diversity (dysbiosis) as compared to healthy controls have been previously reported. Fecal Microbiota Transplantation (FMT) , for the purpose of restoring dysbiosis, has been noted as a novel treatment option for ASD. In this paper, we introduce up‐to‐date observational and intervention studies regarding the microbiota‐gut‐brain axis, and discuss future prospects, such as the possibility of therapeutic application and understanding the pathology of ASD through microbiome studies. There are no potential conflicts of interest to disclose.

Interest in the microbiota‐gut‐brain axis, in which the intestine and the brain interact with each other via the microbiome and its metabolites, is rapidly growing. Autism spectrum disorder (ASD) is one of the most reported microbiota‐related disorders in the neuropsychiatric field. In ASD, differences in specific bacterial species and/or diversity (dysbiosis) as compared to healthy controls have been previously reported. Fecal Microbiota Transplantation (FMT) , for the purpose of restoring dysbiosis, has been noted as a novel treatment option for ASD. In this paper, we introduce up‐to‐date observational and intervention studies regarding the microbiota‐gut‐brain axis, and discuss future prospects, such as the possibility of therapeutic application and understanding the pathology of ASD through microbiome studies. There are no potential conflicts of interest to disclose.

精神疾患の診断は現在，問診に基づく医師の主観的判断によって行われている．このような現在の診断方法は医師の経験に強く依存するため，正確な診断を行うための客観的な診断方法を開発する必要があると言われている．したがって我々の目標は，デバイスによって記録されたデータからうつ病患者の重症度を客観的に計算する深層学習手法を構築することである． 本研究では，うつ病患者と健常者を音声データで分類する深層学習プログラムを開発する．

© 2018 Elsevier B.V. Backgrounds: The intestinal microbiota is considered as a potential common underpinning pathophysiology of Functional Gastrointestinal Disorders (FGIDs) and psychiatric disorders such as depression and anxiety. Fecal Microbiota Transplantation (FMT) has been reported to have therapeutic effects on diseases related to dysbiosis, but few studies have evaluated its effect on psychiatric symptoms. Methods: We followed 17 patients with either Irritable Bowel Syndrome (IBS), Functional Diarrhea (FDr) or Functional Constipation (FC) who underwent FMT for the treatment of gastrointestinal symptoms and observation of psychiatric symptoms. Changes in Hamilton Rating Scale for Depression (HAM-D) and subscale of sleep-related items, Hamilton Rating Scale for Anxiety (HAM-A) and Quick Inventory for Depressive Symptoms (QIDS) between baseline and 4 weeks after FMT, and relationship with the intestinal microbiota were measured. Results: At baseline, 12 out of 17 patients were rated with HAM-D ≥ 8. Significant improvement in HAM-D total and sleep subscale score, HAM-A and QIDS were observed (p = 0.007, p = 0.007, p = 0.01, p = 0.007, respectively). Baseline Shannon index indicated that microbiota showed lower diversity in patients with HAM-D ≥ 8 compared to those of healthy donors and patients with HAM-D < 8. There was a significant correlation between baseline Shannon index and HAM-D score, and a correlation between Shannon index change and HAM-D improvement after FMT. Limitations: The small sample size with no control group. Conclusions: Our results suggest that depression and anxiety symptoms may be improved by FMT regardless of gastrointestinal symptom change in patients with IBS, FDr and FC, and the increase of microbiota diversity may help to improve patient's mood.

Natural language processing (NLP) is a technology in which a computer processes human "natural language" directly. Along with the development of technologies such as automatic morphological analysis and transition words or sentences to vectors, more research that targets psychiatric disorders utilizing NLP is being conducted. In this manuscript, we will review some of these studies. Such approaches may resolve many problems regarding the assessment of psychiatric illnesses, and are expected to contribute to neurolinguistics.

Diagnosis of psychiatric disorders is based primarily on subjective symptoms, and neuroimaging or other biological examinations are used for excluding organic disorders. Advances in artificial intelligence technologies, such as machine learning, may enable us to utilize neuroimaging for individual diagnosis of psychiatric disorder or treatment response prediction. In addition, such technologies may elucidate the underlying pathophysiology of psychiatric disorders. In this article, we review studies that utilized machine learning on structural magnetic resonance imaging for depression.

Long-term benzodiazepines (BZDs) use is not endorsed in the treatment guidelines for anxiety disorders, but is prevalent in the real-world clinical settings. A systematic literature review was performed by using PubMed (last search: May 2019) to identify randomized controlled trials (RCTs) or maintenance studies following RCT that examined the effectiveness of BZDs in patients with anxiety disorders for a duration of 13 weeks or more. Meta-analyses were then conducted regarding changes in the Hamilton Anxiety Rating Scale (HAM-A) scores from baseline through endpoint, all-cause discontinuation, side effects, and the numbers of panic attacks at endpoint. Eight studies were identified (N = 1228). There were no significant differences in all outcomes between BZDs and antidepressants after the initial 8-week treatment. While no significant difference was noted in the HAM-A score changes between BZDs and placebo, BZDs resulted in a lower discontinuation rate and more frequent constipation and dry mouth than placebo. Our study indicates that for those who respond to an initial 8-week treatment, continuing BZDs is equivalent to antidepressants in efficacy and safety. However, the limited number of studies warranted further investigations of the long-term effectiveness and safety of BZDs.