2022 Fiscal Year Final Research Report
Quantification of DAT SPECT in the diagnosis of Parkinson's syndrome and its application to machine learning
Project/Area Number |
19K17243
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 52040:Radiological sciences-related
|
Research Institution | Keio University |
Principal Investigator |
IWABUCHI Yu 慶應義塾大学, 医学部(信濃町), 講師 (90573262)
|
Project Period (FY) |
2019-04-01 – 2023-03-31
|
Keywords | パーキンソン病 / パーキンソン症候群 / ドーパミントランスポーター / DAT SPECT / MIBGシンチグラフィ / 機械学習 / AI / 定量評価 |
Outline of Final Research Achievements |
DAT SPECT, one of the nuclear medicine exam, plays an important role in the diagnosis of Parkinson's syndrome. In this study, we established quantitative evaluation methods to improve the diagnostic performance of DAT SPECT. To date, quantitative evaluation have been used based on the intensity of accumulation of radioisotope in the bilateral striatum and the difference between the left and right sides of the striatum. This study confirmed that a more accurate diagnosis can be performed by evaluating changes in the shape of striatal accumulation. In addition, it was demonstrated that incorporating these multiple quantitative indices into machine learning as feature values and combining them with quantitative values from other nuclear medicine exams such as MIBG scintigraphy enables more comprehensive evaluation and improves diagnostic performance.
|
Free Research Field |
核医学
|
Academic Significance and Societal Importance of the Research Achievements |
本研究成果により核医学検査のひとつであるDAT SPECTによる、より正確かつ客観的なパーキンソン症候群(パーキンソン病含む)の診断体系の確立が出来ると考えられる。また本研究結果は実臨床に直接的に応用可能な内容であり、DAT SPECTでの線条体集積の形態評価を含めた定量解析を実臨床に組み込むことで、パーキンソン症候群の診断能をこれまでより向上させることが出来るという学術的意義があると考える。 パーキンソン症候群をより正確に鑑別診断する事で治療方針の決定や予後の推測などに役立つ臨床情報を得ることができ、さらには予後の改善や介護者の負担軽減にもつながっていくものと思われる。
|