| Project/Area Number |
22K15658
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 51030:Pathophysiologic neuroscience-related
|
|---|
| Research Institution | Institute of Physical and Chemical Research |
|---|
Principal Investigator |
Poon CharissaTingAmanda 国立研究開発法人理化学研究所, 脳神経科学研究センター, 基礎科学特別研究員 (40933130)
|
|---|
| Project Period (FY) |
2022-04-01 – 2026-03-31
|
| Project Status |
Granted (Fiscal Year 2024)
|
| Budget Amount *help |
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2025: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2024: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2023: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2022: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
|
|---|
| Keywords | Alzheimer's disease / image registration / deep learning / segmentation / alzheimer's disease / glial cells / plaques / amyloid / image processing / microscopy / 画像情報処理・画像認識 / 分子・細胞・神経生物学 / 認知症疾患 / グリア細胞 / コンピュータビジョン |
|---|
| Outline of Research at the Start |
Our goal is to create an automated tool to conduct image analyses common in Alzheimer's disease research, specifically: quantification of pathological Aβ plaques and glial cells. Automating analyses will reduce human error and bias, thereby improving the reproducibility of Alzheimer's disease research. The tool will be developed in collaboration with neuroscientists to ensure that it is easily understood and useful.
|
| Outline of Annual Research Achievements |
The goal is to create an interpretable automated image processing pipeline for microscopy images used in Alzheimer's disease research. Due to difficulty in obtaining transgenic animal data, we have been focusing on open datasets, specifically the Seattle Alzheimer's Disease Brain Cell Atlas, which consists of images of human brain tissue. Images of several biomarkers from several brain regions are available for each subject, but the lack of image alignment makes inter-subject comparison difficult. Our current pipeline employs a multi-stage image registration framework, where the first Rigid transformation is calculated using landmarks, which can be refined using deep learning methods. Development has been focused on tissue from the superior temporal and medial temporal brain regions.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
Due to the delay in obtaining microscopy images of transgenic animal data, we have been developing the pipeline using open datasets of images obtained from human AD patients. However, the complexity of the human brain and the limited amount of tissue reflected in each image makes image alignment difficult to achieve. Furthermore, brain sections were likely cut at different angles, which further complicates alignment of brain sections from different individuals.
We are currently focusing on tissue from the superior temporal and medial temporal cortices as these are the most easily visually distinguishable among all the brain regions available in the dataset.
|
| Strategy for Future Research Activity |
Due to continued uncertainty in availability of animal datasets, we plan to proceed development based on the open Seattle Alzheimer's Disease Brain Cell atlas dataset.
The availability of quantitative data in the dataset, such as percentage of the brain occupied by glial biomarkers in different brain regions, is helpful for the validation of the segmentation component of the image processing pipeline.
An ongoing challenge is development and validation of the image registration component of the pipeline, as cropped human brain regions are more difficult to align than whole-brain sections from rodents, from which the original proposal was based on.
Furthermore, the resolution of images in the dataset may not be high enough to see detailed cell structure. As such, the characterization of cell morphology component in the pipeline may be limited. As a first step, we plan to attempt to represent cell shapes using simpler geometric shapes.
|
