| Project/Area Number |
24K21121
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 90130:Medical systems-related
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2024-04-01 – 2027-03-31
|
| Project Status |
Granted (Fiscal Year 2024)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2026: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2025: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2024: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | image segmentation / Predict abnormalities / Abdominal organs / ML algorithms / Segmentation / FDG/PET / CT / Medical imaging / Medical image / Machine learning |
|---|
| Outline of Research at the Start |
Imaging biomarkers of response play a crucial role as an alternative to assessing pathological responses. Using AI, we have been able to segment images accurately. Using this segmented organ's information to predict a patient's future outcome of disease is the main purpose of this research project.
|
| Outline of Annual Research Achievements |
During the first year of this project, we achieved a significant milestone by successfully segmenting fourteen abdominal organs such as the liver, pancreas, spleen, kidneys from computed tomography (CT) images using advanced image processing techniques.
The initial findings from our liver and pancreas analyses were encouraging. We prepared two research abstracts based on these results and submitted them to prominent international conferences in the field of medical imaging and computational diagnostics. We are pleased to report that both submissions were peer-reviewed and accepted for presentation. These recognitions underscore the scientific value and potential impact of our work.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
As a core component of this project, we successfully completed the segmentation of abdominal organs in a total of 8,850 CT images. This segmentation task is fundamental to the project's goals, as it enables organ-specific analysis such as volumetric assessment, density measurements, and correlation with clinical parameters. The completion of this step marks a major achievement and lays the groundwork for downstream analyses involving disease prediction, organ health assessment, and population-based studies.
In parallel with this technical accomplishment, we also began exploring and analyzing organ-specific patterns; particularly focusing on the liver and pancreas. Based on this work, we prepared and submitted abstracts to two reputable international conferences.
|
| Strategy for Future Research Activity |
Our focus this year has shifted toward the discovery of imaging biomarkers that can predict early signs of organ-specific abnormalities. We are leveraging state-of-the-art machine learning and statistical tools to extract meaningful features from the segmented data, with the goal of identifying biomarkers that correlate with metabolic, inflammatory, or structural changes associated with disease.
Looking ahead, we plan to conduct comprehensive biomarker analyses for all fourteen segmented organs. This will allow us to construct a multi-organ risk profile for individuals and explore inter-organ relationships in the context of health and disease.
|
