WEKO3
アイテム
Parametric Imaging of the Total Volume of Distribution using MAP Estimation for Logan Graphical Analysis
https://repo.qst.go.jp/records/69425
https://repo.qst.go.jp/records/69425a7730bce-5c48-4962-82df-6c62cbe73e3c
| Item type | 会議発表用資料 / Presentation(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2008-08-06 | |||||
| タイトル | ||||||
| タイトル | Parametric Imaging of the Total Volume of Distribution using MAP Estimation for Logan Graphical Analysis | |||||
| 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 |
Shidahara, Miho
× Shidahara, Miho× Kimura, Yuichi× Seki, Chie× Naganawa, Mika× Sakata, Muneyuki× Ito, Hiroshi× Suhara, Tetsuya× Ishiwata, Kiichi× Kanno, Iwao× et.al× 志田原 美保× 木村 裕一× 関 千江× 長縄 美香× 坂田 宗之× 伊藤 浩× 須原 哲也× 石渡 喜一× 菅野 巖 |
|||||
| 抄録 | ||||||
| 内容記述タイプ | Abstract | |||||
| 内容記述 | Introduction: We propose MAP estimation algorithm in graphical analysis (MEGA) to reduce noise-induced bias of the total volume of distribution (VT) images estimated by Logan graphical analysis (GA) [1]. The purpose of this study is to evaluate its applicability using a s1 receptor ligand 11C-SA4503 and a dopamine transporter ligand 11C-PE2I. Methods: In general, MAP estimation calculates a parameter s while maximization of P(D|s)P(s), where P(D|s) and P(s) are likelihood of datasets D and prior probability of s, respectively. In MEGA, the Mahalanobis distance in a feature space was utilized for the likelihood term in MAP, and a uniform distribution between given lower and upper bounds was applied for the prior. Practically, a set of noise free time-activity curves (TACs) was formed with VT and the y-intercept in GA varying in physiological range as a template, and then the most similar TAC was sought out for a given measured TAC in a feature space. The Similarities in shape were measured using the Mahalanobis distance. In simulation, MEGA was compared with other three methods: GA, likelihood estimation in GA (LEGA) [2] and Multilinear analysis (MA1) [2] using 500 noisy TACs under seven physiological conditions (from 9.9 to 61.5 of VT). PET studies of 11C-SA4503 (at baseline and after loading of Fluvoxamine for partial blockade of the s1 receptors) and 11C-PE2I (at baseline) were performed for three normal volunteers and for a normal volunteer, respectively. The VT images of 11C-SA4503 estimated by MEGA were compared with ROI estimates by GA over 4 brain regions. For 11C-PE2I, an estimated VT image by MEGA was also compared against ROI estimates by GA in 12 high and low binding regions. Results: In the simulation study, the estimated VT by GA had large underestimation (y = 0.27x + 8.72, r2 = 0.87; x is true and y is a median of estimates). Applying the other methods (MA1, LEGA, and MEGA), this bias was improved (y = 0.80x + 4.04, r2 = 0.98; y = 0.85x + 3.05, r2 = 0.99; y = 0.96x + 1.21, r2 = 0.99, respectively). MA1 and LEGA increased variance of the estimated VT in simulation and clinical VT images. However, visual improvements of VT images estimated by MEGA were observed as shown in Figure 1. MEGA was proportional to ROI-based estimated VT: y = 0.81x + 5.5 (r2 = 0.95) for 11CSA4503 and y = 1.05x + 0.28 (r2 = 1.0) for 11C-PE2I, respectively. Conclusion: MEGA could improve estimates of VT in clinical neuroreceptor PET imaging. |
|||||
| 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 | NeuroReceptor Mapping 2008 | |||||
| 発表年月日 | ||||||
| 日付 | 2008-07-19 | |||||
| 日付タイプ | Issued | |||||
