Institute of One is the research division of LISIT Co., Ltd. It publishes open, reproducible research in medical imaging physics — CT dosimetry, radiomics and image quality — and releases the code and archived data behind every result.

Affiliation
Institute of One, LISIT Co., Ltd., Tokyo 150-0044, Japan
Principal investigator
Shuji Yamamoto, PhD
ORCID: 0000-0001-9211-1071
Contact: [email protected]

Peer-reviewed publications

A Stability Atlas for IBSI Radiomics Features Using Synthetic Digital Phantoms, with Proof-of-Concept Physics-Based Normalisation
Journal of Imaging 12(8), 392 (2026) · CC BY 4.0
Anatomy-Weighted CTDIvol from Routine CT Metadata: A Patient-Specific, Multi-Vendor Study Using Deep-Learning Segmentation
Tomography 12(9), 125 (2026) · CC BY 4.0 · published 30 August 2026

Preprint and manuscripts under submission

An Open, Reproducible Gamma-Variate Pipeline for CT-Perfusion Curve Analysis
Preprint: medRxiv, 26 June 2026 · submitted to SoftwareX, September 2026
An Open, Closed-Form-Validated Framework for Task-Based Image Quality on Synthetic Phantoms
Under review at Journal of Imaging; revision submitted September 2026
Denoising under a Data-Processing Ceiling: Observer-Dependent Benefits, Fidelity–Task Divergence, and an Information Floor
Submitted to Medical Physics, August 2026
A Measured Floor on Index-Based Organ Dose Estimation in CT, Where Per-Organ Coefficients Do Not Improve on a Single Conversion Factor
Submitted to Physics in Medicine & Biology, September 2026
Sampling Design Bounds Parameter Recovery in a Reduced CT Contrast-Kinetics Model, and a Closed-Form Fit Nearly Attains the Bound
Submitted to Computer Methods and Programs in Biomedicine, August 2026
Detection Information Saturates over a Quarter of the Reconstruction Band, and the Missing Band Is Lost in the Observer rather than the Display
Submitted to Medical Physics, September 2026

In preparation, and completed work not submitted

An Open-Source, Provenance-Aware Framework for Reconstructing Missing CT Dose Indices from DICOM Metadata
Revised Technical Note in preparation for the Journal of Applied Clinical Medical Physics
Topology and detectability in medical images
Complete and not submitted. All five matrix sizes were measured against a preregistered outcome: a reproducible nonlinear topological crossover accompanies the loss of ideal-observer detectability, but its location depends on matrix and lesion size and does not collapse onto detectability alone. The study was set aside on scope rather than on its result, and the record is kept public so the negative finding remains available.

Every study above ships its implementation as an openly licensed repository. Nine repositories are permanently archived with DOIs, all MIT-licensed and all citable; two more are public while their work continues — ldct-io, the reader and reconstruction package for measured CT projection data, and topology-img-core.

Full research record: instituteofone.org