2022 Fiscal Year Final Research Report
Radiomics analysis of breast images for precision medicine
| Project/Area Number |
20K08131
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 52040:Radiological sciences-related
|
|---|
| Research Institution | Shiga University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
大岩 幹直 独立行政法人国立病院機構(名古屋医療センター臨床研究センター), その他部局等, 医長 (50649697)
川崎 朋範 埼玉医科大学, 医学部, 教授 (90456484)
|
|---|
| Project Period (FY) |
2020-04-01 – 2023-03-31
|
| Keywords | breast cancer / precision medicine / radiomics / subtype classification / deep learning |
|---|
| Outline of Final Research Achievements |
The purpose of this study was to analyze breast cancer diagnostic images for classification of cancer subtypes and histological grades to assist radiologists in diagnosis and treatment planning and to contribute to precision medicine. For developing such systems, high quality database that includes multimodality images and the histologic information from multi-centers is required. In this study, we collected 600 of such cases. For classification of subtypes and histological grades, we compared single modality models and multimodality models and confirmed the higher classification accuracy with the multimodality models.
|
| Free Research Field |
medical image analysis
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では乳がんの診断の初期に用いられる画像によりがんのサブタイプの予測を行い，診断にかかる時間の短縮とよりスムーズな治療計画の決定により患者の経済的かつ心理的負担軽減を試みた．まだ予測精度は十分ではないが，本研究により診断画像によるサブタイプ分類の可能性が示唆された．本研究により，この分野の研究が更に進み，今後予測精度が向上すれば，乳がんの最適化医療への貢献が期待できる．
|
