2023 Fiscal Year Final Research Report
Imaging biopsy connecting detection of genetic mutation with prognostic prediction
| Project/Area Number |
20K08084
|
|---|
| 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 | Kyushu University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
田中 謙太郎 九州大学, 大学病院, 助教 (00536849)
|
|---|
| Project Period (FY) |
2020-04-01 – 2024-03-31
|
| Keywords | 画像生検 / 穴解析 / 肺癌 / 位相幾何学 / ベッチ数マップ |
|---|
| Outline of Final Research Achievements |
The theory focused on in this study is topology, which is called a soft geometry, a theory that classifies objects by the number of holes without regard to their concrete shapes. In this study, we call it "hole analysis" [Hidetaka Arimura et al. Medical Imaging and Information Sciences 2023]. Since lung cancers with EGFR mutations tend to show cavities and bubbles on images, we hypothesized that the hole analysis would be feasible for detection of EGFR mutations. In this study, we used Betti number map features to identify image invariant features corresponding to EGFR mutations and developed a noninvasive image biopsy with the hole analysis. We demonstrated that the Betti number map features based on topology can detect lung cancer with EGFR mutations (including subtype classification of Del19/L858R mutations), which is the hypothesis we proved in this study.
|
| Free Research Field |
医用画像情報学、医学物理学
|
| Academic Significance and Societal Importance of the Research Achievements |
位相幾何学に基づくベッチ数マップ特徴量による予測モデルは、非小細胞肺癌患者におけるEGFR変異の同定において、CT装置／撮像パラメータの変動に対して従来モデルよりも高い頑健性を示した。さらに、EGFR Del19/L858R変異のサブタイプの特徴と遺伝学的関連を示した3Dベッチ数マップ特徴量は、従来特徴量と比較してサブタイプ分類に高い精度を示した。
現状の生検では、不均一な遺伝子変異を持つ腫瘍の遺伝子変異予測を過小または過大評価する可能性があるが、この問題を提案する画像生検が解決できる可能性がある。また、生検不可または拒否の患者に対して画像生検を適用できる。
|
