WEKO3
アイテム
Identification of antidepressant dose-related, resting-state functional connectivity as a novel therapeutic target in neurofeedback: a machine learning-based fMRI study
https://repo.qst.go.jp/records/72573
https://repo.qst.go.jp/records/72573cab3c4b6-2618-4c6a-aac0-fdc9f749825a
| Item type | 会議発表用資料 / Presentation(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2017-12-13 | |||||
| タイトル | ||||||
| タイトル | Identification of antidepressant dose-related, resting-state functional connectivity as a novel therapeutic target in neurofeedback: a machine learning-based fMRI study | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_c94f | |||||
| 資源タイプ | conference object | |||||
| アクセス権 | ||||||
| アクセス権 | metadata only access | |||||
| アクセス権URI | http://purl.org/coar/access_right/c_14cb | |||||
| 著者 |
Yahata, Noriaki
× Yahata, Noriaki× Ichikawa, Naho× Lisi, Giuseppe× Morimoto, Jun× Okamoto, Yasumasa× Kawato, Mitsuo× 八幡 憲明 |
|||||
| 抄録 | ||||||
| 内容記述タイプ | Abstract | |||||
| 内容記述 | Recent studies have indicated that resting-state functional connectivity (rs-fc) holds great promise for effectively delineating the disruption in the neural circuits caused by mental disorders. By applying machine learning techniques to mass rs-fc MRI data, we have earlier identified a small number of functional connections (FCs) that reliably distinguished healthy controls from patients with mental disorders such as autism spectrum disorder (ASD) and major depressive disorder (MDD). However, the extent to which the identified FCs were influenced by the administration of psychotropic drugs in the patients (e.g. antidepressant to treat MDD patients) is uncertain, making it difficult to evaluate the net effect of mental disorders on a particular FC. To approach this question, here we conducted a machine learning study to identify FCs associated with the administration of selective serotonin reuptake inhibitor (SSRI), a first-line antidepressant to treat MDD patients. We then compared the results with the independent set of FCs that reliably predicted the diagnostic status (MDD or healthy) of each individual [3]. A machine learning algorithm, developed previously to construct an FC-based classifier for ASD, was applied to a data set consisting of MDD patients (N=82 with SSRI and N=22 without SSRI) and healthy controls (N=143) in order to extract FCs specifically related to the status of SSRI administration. The algorithm identified a total of 23 SSRI dose-specific FCs distributed across the whole brain, by which the patients with and without SSRI treatment were successfully distinguished (area under the curve, AUC=0.80). The identified FCs did not overlap with the set of FCs that predicted the diagnostic status of an individual. Furthermore, this reliability of the classification was generalized to an independent cohort that consisted of individuals with ASD (N=29 with SSRI and N=45 without SSRI treatment) and typically-developed controls (N=107) (AUC=0.73). The present study suggests that the effects of MDD pathophysiology and SSRI treatment on FCs can be identified and evaluated separately. It is also suggested that the SSRI dose-related FCs may be a novel therapeutic target to treat MDD patients through neurofeedback, especially those who present poor drug compliance. | |||||
| 会議概要（会議名, 開催地, 会期, 主催者等） | ||||||
| 内容記述タイプ | Other | |||||
| 内容記述 | Real-time Functional Imaging and Neurofeedback Conference 2017 | |||||
| 発表年月日 | ||||||
| 日付 | 2017-11-30 | |||||
| 日付タイプ | Issued | |||||
