※ Affiliations and titles are as of the end of the research activity.

Masahiro Ando

Comprehensive Molecular Imaging of Living Cells by Raman spectroscopy and Sparse Analysis

Researcher

Masahiro Ando

Outline

Label-free and Non-destructive molecular-level analysis of living cells is essential to all kinds of cell-based biotechnology and medicine. Raman spectroscopy is one of the promising tools for visualizing comprehensive molecular distributions in living cells. In this study, overcoming difficulties in interpreting complex and low S/N spectra, information science algorithms of sparse analysis are effectively coalesced to achieve automatic and comprehensive molecular imaging of biological systems.

Hiroki Ogawa

Development of visualizing method through cooperative small angle x-ray scattering coupled with computed tomography (SAXS-CT) and information science

Researcher

Hiroki Ogawa

Outline

As an measument technique, SAXS-CT method has been used for analyzing self-assembled morphologies of softmaterials. However, only the approach from the measurement technology, this application have limited uses to softmaterials. In cllaborated with information science, I will develop a new SAXS-CT method to visualize nanostructural information.

Shunsuke Ono

Signal Reconstruction with Decomposition by Unified Convex Optimization Framework

Researcher

Shunsuke Ono

Outline

This project aims to establish a flexible framework for reconstructing the signal of interest from degraded observations. The key idea of our framework is twofold. One is to model the signal of interest as the sum of a number of meaningful components and reconstruct them in parallel with component-wise regularizations that exploit a-priori information on each component. This enables high-quality signal reconstruction in low signal-to-noise ratio scenarios. The other is to develop an efficient and scalable convex optimization methodology that can solve various signal reconstruction problems in a unified manner.

Takuma Kasai

Advanced NMR analysis with information-encoded samples, integrated measurements, and tensor decomposition

Researcher

Takuma Kasai

Outline

Nuclear magnetic resonance (NMR) is a powerful method to investigate structures and dynamics of proteins at atomic resolution. However, the application of NMR for large proteins or proteins in living cells is limited due to enhanced relaxation and signal overlap. To overcome these difficulties, I integrate several new approaches: elaborated isotope labeling scheme to encode information for assignments, combined NMR measurements for simultaneous assignments and analyses, tensor decomposition for signal separation. This strategy will help application of NMR for various proteins in challenging situations.

Kenji Katayama

Development of modulated light inducd phase microscopy and evaluation for lifetime and transport property of photoexcited carriers of photo-devices

Researcher

Kenji Katayama

Outline

There are some key photo-devices in near future, such as solar cell, photocatalyst, and artificial photosynthesis, where photo-excited carriers (electrons and holes) are extracted for specific purpuses. In these devices, assembly of semiconductor particles are used as electrodes. However, most of the carriers are usually trapped to defects, and their lifetime and spatial distribution are diversified, which have not been understood well. In this study, we will develop a new method which can clarify the correlation between the arrangement of particles and the carrier properties, especially on the lifetime and the transport property.

Tatsu Kuwatani

Extraction of geodynamic processes from rocks: Bayesian sensing challenges the ultimate inverse problem

Researcher

Tatsu Kuwatani

Outline

Rocks preserve precious information about seismicity, volcanic activity and resource formation. One of the most important instruments in petrology is the Electron Probe MicroAnalyzer (EPMA), electron microscope which can measure elemental compositions. By the Bayesian sensing, which combines measurement technologies with information science, this study tries to maximize the performance of EPMA, and challenges the ultimate inverse problem of extracting all related geodynamic processes from rocks.

Tomoya Nakamura

Ultra-compact computational camera by using artificial scattering media and compressive sensing

Researcher

Tomoya Nakamura

Outline

The objective of this proposal is the invention of a novel design of the computational lensless camera, which can capture the large amount of image data with an ultra-compact hardware. Specifically, this proposal aims to realize omni-directional and multi-spectral imaging with a small optical hardware whose size is the order of mm^3. The goal of this study is the verification of such imaging capability for the application of the “optical big-data sensor”, and the creation of a new field for the image industry in the era of information technology. The possible application includes the cognitive capsule endoscope, the smart robot with the ultra-compact vision system, the ultra-light wearable sensor, and so on.

Taku Nonomura

Development of Fast and Accurate Data Assimilation towards Optimal Flow Control

Researcher

Taku Nonomura

Outline

Towards the optimal flow control, series of techniques for fast and accurate data assimilation of flow fields are investigated and developed. Following items are studied: :(i) Acquisition of the detailed unsteady data by optical flow technique and reduced order modeling, and (ii) new method based on calculus of variation with the sub-optimal control theory. With regard to the item (i), detailed unsteady flow data is obtained and the reduced order modeling is applied for reduction of computational costs. With regard to the item (ii) the sub-optimal theory is utilized for the construction of fast and accurate data assimilation.

Yasuhiro Matsunaga

Development and Application of Data Assimilation Methods for Biomolecules

Researcher

Yasuhiro Matsunaga

Outline

Data assimilation is a statistical method designed to improve the quality of numerical simulations in combination with real observations. In order to characterize the dynamical processes of biomolecules when they function, we develop a new data assimilation method for biomolecular simulations and experiments. Through the applications of the method to real experimental data, folding pathways and conformational changes of proteins are investigated in detail. We also generalize the method to integrate various types of experimental data including from single-molecule measurements to bulk measurements.

Yoshihiro Watanabe

High-speed sensing of reflecatnce properties based on sparse modeling and dynamic light-field control

Researcher

Yoshihiro Watanabe

Outline

The purpose of this research is to realize a high-speed and high-quality sensing technology for reflectance properties. This type of sensing has taken a considerable time because such data has multidimensional information space. If this bottleneck can be removed, the progress can contribute to various applications including digital archiving, production, inspection, video production, and so on. Based on s.uch background, this research realizes the technologies to capture reflectance properties with a minimum measurement time and with keeping high quality based on sparse modeling and dynamic light-field control.