핵의학

본문글자크기
  • [Eur J Nucl Med Mol Imaging.] Prediction of post-stroke cognitive impairment using brain FDG PET: deep learning-based approach 딥러닝기반 FDG PET으로 뇌졸중 후 인지장애를 예측

    중앙의대 / 이리리, 최홍윤, 박광열*, 석주원*

  • 출처
    Eur J Nucl Med Mol Imaging.
  • 등재일
    2022 Mar
  • 저널이슈번호
    49(4):1254-1262. doi: 10.1007/s00259-021-05556-0.
  • 내용

    바로가기  >

    Abstract
    Purpose: Post-stroke cognitive impairment can affect up to one third of stroke survivors. Since cognitive function greatly contributes to patients' quality of life, an objective quantitative biomarker for early prediction of dementia after stroke is required. We developed a deep-learning (DL)-based signature using positron emission tomography (PET) to objectively evaluate cognitive decline in patients with stroke.

    Methods: We built a DL model that differentiated Alzheimer's disease (AD) from normal controls (NC) using brain fluorodeoxyglucose (FDG) PET from the Alzheimer's Disease Neuroimaging Initiative database. The model was directly transferred to a prospectively enrolled cohort of patients with stroke to differentiate patients with dementia from those without dementia. The accuracy of the model was evaluated by the area under the curve values of receiver operating characteristic curves (AUC-ROC). We visualized the distribution of DL-based features and brain regions that the model weighted for classification. Correlations between cognitive signature from the DL model and clinical variables were evaluated, and survival analysis for post-stroke dementia was performed in patients with stroke.

    Results: The classification of AD vs. NC subjects was performed with AUC-ROC of 0.94 (95% confidence interval [CI], 0.89-0.98). The transferred model discriminated stroke patients with dementia (AUC-ROC = 0.75). The score of cognitive decline signature using FDG PET was positively correlated with age, neutrophil-lymphocyte ratio and platelet-lymphocyte ratio and negatively correlated with body mass index in patients with stroke. We found that the cognitive decline score was an independent risk factor for dementia following stroke (hazard ratio, 10.90; 95% CI, 3.59-33.09; P < 0.0001) after adjustment for other key variables.

    Conclusion: The DL-based cognitive signature using FDG PET was successfully transferred to an independent stroke cohort. It is suggested that DL-based cognitive evaluation using FDG PET could be utilized as an objective biomarker for cognitive dysfunction in patients with cerebrovascular diseases.

     

     

    Affiliations

    Reeree Lee #  1 , Hongyoon Choi #  2 , Kwang-Yeol Park  3 , Jeong-Min Kim  4 , Ju Won Seok  5
    1 Department of Nuclear Medicine, Chung-Ang University Hospital, Chung-Ang University College of Medicine, 224-1, Heukseok-dong, Dongjak-gu, Seoul, 06974, Republic of Korea.
    2 Department of Nuclear Medicine, Seoul National University Hospital, Seoul, Republic of Korea.
    3 Department of Neurology, Chung-Ang University Hospital, Chung-Ang University College of Medicine, 224-1, Heukseok-dong, Dongjak-gu, Seoul, 06974, Republic of Korea. kwangyeol.park@gmail.com.
    4 Department of Neurology, Seoul National University Hospital, Seoul, Republic of Korea.
    5 Department of Nuclear Medicine, Chung-Ang University Hospital, Chung-Ang University College of Medicine, 224-1, Heukseok-dong, Dongjak-gu, Seoul, 06974, Republic of Korea. joneseok@cau.ac.kr.
    # Contributed equally.

  • 키워드
    Deep learning; Post-stroke cognitive impairment; Post-stroke dementia; [18F]FDG.
  • 편집위원

    뇌경색이후 발생하는 인지장애를 18F-FDG PET CT 영상에서 deep-learning 기법을 이용하여 예측하는 임상연구임. 해당 연구를 통해 18F-FDG PET CT 영상의 deep-learning 평가가 뇌경색후 발생되는 인지장애를 예측할 수 있는 바이오마커임을 확인함. 해당 연구는 혈관성뇌질환 및 인지장애에 관심이 많은 임상과와 신경핵의학 및 인공지능 개발 연구자에게 관심을 끌 흥미로운 연구로 생각됨.

    2022-05-04 16:30:48

  • 편집위원2

    딥러닝을 활용해서 이미지 데이터를 분석하고, 생존분석을 활용해서 기존의 Hazard ratio를 각각 분석하였다. 추후에는 이미지데이터와 기타 risk factor 의 데이터를 연계해서, 머신러닝 기법과 통계기법을 활용할 수 있는 아이디어를 제공해주는 사전 논문으로써 가치가 있다고 생각한다.

    2022-05-04 16:50:10

  • 덧글달기
    덧글달기
       IP : 3.215.79.204

    등록