Machine Learning Model to Predict Diagnosis of Mild Cognitive Impairment by Using Radiomic and Amyloid Brain PET

被引:7
作者
Ciarmiello, Andrea [1 ]
Giovannini, Elisabetta [1 ]
Pastorino, Sara [1 ]
Ferrando, Ornella [2 ]
Foppiano, Franca [2 ]
Mannironi, Antonio [3 ]
Tartaglione, Antonio [4 ]
Giovacchini, Giampiero [1 ]
机构
[1] S Andrea Hosp, Nucl Med Unit, La Spezia, Italy
[2] S Andrea Hosp, Med Phys Unit, La Spezia, Italy
[3] S Andrea Hosp, Neurol Unit, La Spezia, Italy
[4] Memory Lab CNS ONLUS, La Spezia, Italy
关键词
aMCI; amyloid PET; neural network; ALZHEIMERS-DISEASE; TEXTURE ANALYSIS; IMAGES; CONVERSION; BIOMARKER;
D O I
10.1097/RLU.0000000000004433
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
PurposeThe study aimed to develop a deep learning model for predicting amnestic mild cognitive impairment (aMCI) diagnosis using radiomic features and amyloid brain PET.Patients and MethodsSubjects (n = 328) from the Alzheimer's Disease Neuroimaging Initiative database and the EudraCT 2015-001184-39 trial (159 males, 169 females), with a mean age of 72 +/- 7.4 years, underwent PET/CT with F-18-florbetaben. The study cohort consisted of normal controls (n = 149) and subjects with aMCI (n = 179). Thirteen gray-level run-length matrix radiomic features and amyloid loads were extracted from 27 cortical brain areas. The least absolute shrinkage and selection operator regression was used to select features with the highest predictive value. A feed-forward neural multilayer network was trained, validated, and tested on 70%, 15%, and 15% of the sample, respectively. Accuracy, precision, F1-score, and area under the curve were used to assess model performance. SUV performance in predicting the diagnosis of aMCI was also assessed and compared with that obtained from the machine learning model.ResultsThe machine learning model achieved an area under the receiver operating characteristic curve of 90% (95% confidence interval, 89.4-90.4) on the test set, with 80% and 78% for accuracy and F1-score, respectively. The deep learning model outperformed SUV performance (area under the curve, 71%; 95% confidence interval, 69.7-71.4; 57% accuracy, 48% F1-score).ConclusionsUsing radiomic and amyloid PET load, the machine learning model identified MCI subjects with 84% specificity at 81% sensitivity. These findings show that a deep learning algorithm based on radiomic data and amyloid load obtained from brain PET images improves the prediction of MCI diagnosis compared with SUV alone.
引用
收藏
页码:1 / 7
页数:7
相关论文
共 48 条
[1]  
ADNI, 2008, DEF ALZH DIS PROC MA
[2]   Cerebral amyloid-β PET with florbetaben (18F) in patients with Alzheimer's disease and healthy controls: a multicentre phase 2 diagnostic study [J].
Barthel, Henryk ;
Gertz, Hermann-Josef ;
Dresel, Stefan ;
Peters, Oliver ;
Bartenstein, Peter ;
Buerger, Katharina ;
Hiemeyer, Florian ;
Wittemer-Rump, Sabine M. ;
Seibyl, John ;
Reininger, Cornelia ;
Sabri, Osama .
LANCET NEUROLOGY, 2011, 10 (05) :424-435
[3]   Diagnostic and prognostic value of amyloid PET textural and shape features: comparison with classical semi-quantitative rating in 760 patients from the ADNI-2 database [J].
Ben Bouallegue, Faycal ;
Vauchot, Fabien ;
Mariano-Goulart, Denis ;
Payoux, Pierre .
BRAIN IMAGING AND BEHAVIOR, 2019, 13 (01) :111-125
[4]   Experimental phantom evaluation to identify robust positron emission tomography (PET) radiomic features [J].
Carles, Montserrat ;
Fechter, Tobias ;
Marti-Bonmati, Luis ;
Baltas, Dimos ;
Mix, Michael .
EJNMMI PHYSICS, 2021, 8 (01)
[5]   Effects of Socioeconomic Status on Alzheimer Disease Mortality in Taiwan [J].
Chen, Nai-Ching ;
Liang, Chih-Kuang ;
Yin, Chun-Hao ;
Lin, Yu-Te ;
Lee, Ching-Chih ;
Chen, Chien-Liang .
AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY, 2020, 28 (02) :205-216
[6]   Cognitive signature of brain FDG PET based on deep learning: domain transfer from Alzheimer's disease to Parkinson's disease [J].
Choi, Hongyoon ;
Kim, Yu Kyeong ;
Yoon, Eun Jin ;
Lee, Jee-Young ;
Lee, Dong Soo .
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2020, 47 (02) :403-412
[7]   Predicting cognitive decline with deep learning of brain metabolism and amyloid imaging [J].
Choi, Hongyoon ;
Jin, Kyong Hwan .
BEHAVIOURAL BRAIN RESEARCH, 2018, 344 :103-109
[8]   Longitudinal cognitive decline in mild cognitive impairment subjects with early amyloid-β neocortical deposition [J].
Ciarmiello, Andrea ;
Giovannini, E. ;
Riondato, M. ;
Giovacchini, G. ;
Duce, V. ;
Ferrando, O. ;
De Biasi, M. ;
Passera, C. ;
Carabelli, E. ;
Mannironi, A. ;
Mansi, L. ;
Tartaglione, A. .
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2019, 46 (10) :2090-2098
[9]   Amyloid burden identifies neuropsychological phenotypes at increased risk of progression to Alzheimer's disease in mild cognitive impairment patients [J].
Ciarmiello, Andrea ;
Tartaglione, Antonio ;
Giovannini, Elisabetta ;
Riondato, Mattia ;
Giovacchini, Giampiero ;
Ferrando, Ornella ;
De Biasi, Marina ;
Passera, Chiara ;
Carabelli, Elena ;
Mannironi, Antonio ;
Foppiano, Franca ;
Alfano, Bruno ;
Mansi, Luigi .
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2019, 46 (02) :288-296
[10]   Radiomic Features of Hippocampal Subregions in Alzheimer's Disease and Amnestic Mild Cognitive Impairment [J].
Feng, Feng ;
Wang, Pan ;
Zhao, Kun ;
Zhou, Bo ;
Yao, Hongxiang ;
Meng, Qingqing ;
Wang, Lei ;
Zhang, Zengqiang ;
Ding, Yanhui ;
Wang, Luning ;
An, Ningyu ;
Zhang, Xi ;
Liu, Yong .
FRONTIERS IN AGING NEUROSCIENCE, 2018, 10