Presentation | 2022-12-23 Selective Inference for Cluster Level Inference in Brain Image Analysis Masaya Ikuta, Mizuki Sato, Akifumi Yamada, Vo Nguyen Le Duy, Ryo Emoto, Yuko Ishimaru, Yuka Takao, Atsushi Kawaguchi, Shigeyuki Matsui, Ichiro Takeuchi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Cluster-level inference in brain image analysis is often employed to identify disease-related regions in brain disorders. Cluster-level inference is a statistical analysis method in which a brain image is divided into the smallest units called voxels, and the disease-related regions is represented as a set of voxels (cluster) based on the assigned values to voxels. In this study, we consider the hypothesis testing problem for clusters as inference for data-driven hypotheses and propose a conditional selective inference (CSI) framework for brain-image cluster-level inference. CSI is a framework for performing inference conditional on the selection of hypotheses, which allows us to devise a valid statistical test that can control the false positive rate to a specified level of significance. We demonstrate the validity of the proposed method by demonstrating that the false positive rate can be properly controlled in numerical experiments. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | conditional SI / data-driven hypothesis / brain image analysis / cluster level inference |
Paper # | IBISML2022-61 |
Date of Issue | 2022-12-15 (IBISML) |
Conference Information | |
Committee | IBISML |
---|---|
Conference Date | 2022/12/22(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Kyoto University |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Machine Learning, etc. |
Chair | Masashi Sugiyama(Univ. of Tokyo) |
Vice Chair | Toshihiro Kamishima(AIST) / Koji Tsuda(Univ. of Tokyo) |
Secretary | Toshihiro Kamishima(NTT) / Koji Tsuda(Hokkaido Univ.) |
Assistant | Yoshinobu Kawahara(Osaka Univ.) / Taiji Suzuki(Tokyo Inst. of Tech.) |
Paper Information | |
Registration To | Technical Committee on Infomation-Based Induction Sciences and Machine Learning |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Selective Inference for Cluster Level Inference in Brain Image Analysis |
Sub Title (in English) | |
Keyword(1) | conditional SI |
Keyword(2) | data-driven hypothesis |
Keyword(3) | brain image analysis |
Keyword(4) | cluster level inference |
1st Author's Name | Masaya Ikuta |
1st Author's Affiliation | Nagoya Institute of Technology(NITech) |
2nd Author's Name | Mizuki Sato |
2nd Author's Affiliation | Nagoya Institute of Technology(NITech) |
3rd Author's Name | Akifumi Yamada |
3rd Author's Affiliation | Nagoya University(Nagoya Univ.) |
4th Author's Name | Vo Nguyen Le Duy |
4th Author's Affiliation | Nagoya Institute of Technology/RIKEN(NITech/RIKEN) |
5th Author's Name | Ryo Emoto |
5th Author's Affiliation | Nagoya University(Nagoya Univ.) |
6th Author's Name | Yuko Ishimaru |
6th Author's Affiliation | Saga University(Saga Univ.) |
7th Author's Name | Yuka Takao |
7th Author's Affiliation | Saga University(Saga Univ.) |
8th Author's Name | Atsushi Kawaguchi |
8th Author's Affiliation | Saga University(Saga Univ.) |
9th Author's Name | Shigeyuki Matsui |
9th Author's Affiliation | Nagoya University(Nagoya Univ.) |
10th Author's Name | Ichiro Takeuchi |
10th Author's Affiliation | Nagoya University/RIKEN(Nagoya Univ./RIKEN) |
Date | 2022-12-23 |
Paper # | IBISML2022-61 |
Volume (vol) | vol.122 |
Number (no) | IBISML-325 |
Page | pp.pp.128-133(IBISML), |
#Pages | 6 |
Date of Issue | 2022-12-15 (IBISML) |