{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T18:04:39Z","timestamp":1784657079112,"version":"3.55.0"},"reference-count":47,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T00:00:00Z","timestamp":1784592000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000181","name":"Air Force Office of Scientific Research","doi-asserted-by":"publisher","award":["FA9550-22-1-0272"],"award-info":[{"award-number":["FA9550-22-1-0272"]}],"id":[{"id":"10.13039\/100000181","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS-2024-00406152"],"award-info":[{"award-number":["RS-2024-00406152"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS-2023-00301938"],"award-info":[{"award-number":["RS-2023-00301938"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Distinguishing individuals with cognitive decline (CD), including early Alzheimer\u2019s disease, from cognitively normal (CN) individuals is essential for improving diagnostic accuracy and enabling timely intervention. Positron emission tomography (PET) captures metabolic brain alterations associated with CD, but its broader application is often limited by cost and radiation exposure. To enhance the clinical utility of PET while addressing data limitations, we propose a data-efficient framework that integrates complementary multi-scale PET representations at voxel-level and region-level.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Voxel-level features were extracted using convolutional neural networks (CNN) or principal component analysis networks (PCANet) from [\u00b9\u2078F]FDG PET imaging. Region-level features were derived from standardized uptake value ratio measurements across predefined brain regions and processed using a deep neural network (DNN). These voxel- and region-level information are integrated through direct concatenation. For the final prediction, different machine learning models and ensemble technique were applied. The models were trained and validated using 5-fold cross-validation on PET scans from 252 participants in the Alzheimer\u2019s Disease Neuroimaging Initiative, comprising 118 CN and 134 CD subjects. Additional correlation analysis and disease classification comparison with the Mini-Mental State Examination (MMSE) were also performed.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In 5-fold cross-validation, CNN, PCANet, and DNN models achieved classification accuracies of 0.69\u2009\u00b1\u20090.04, 0.69\u2009\u00b1\u20090.06, and 0.82\u2009\u00b1\u20090.06, respectively. The integrated DNN-CNN model using direct concatenation yielded the highest accuracy (0.87\u2009\u00b1\u20090.05), with a 6.33% improvement in accuracy and reduced standard deviation relative to the DNN-only model. Overall, there were an increase of 14.22% in Recall (0.77 to 0.88) and an increase of 7.92% in F1-Score (0.82 to 0.88). Moreover, the predicted probability of CD showed a significant correlation with MMSE scores, and the model achieved higher accuracy, recall, and F1-score than MMSE-based classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Combining complementary voxel-level and region-level PET representations with deep learning improved classification performance over single-representation models, particularly by enhancing sensitivity to cognitive decline. These findings support the potential utility of multi-scale FDG-PET representations for machine learning-based cognitive decline detection.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0341995","type":"journal-article","created":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T17:23:38Z","timestamp":1784654618000},"page":"e0341995","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Feature integration of [18F]FDG PET brain imaging using deep learning for sensitive cognitive decline detection"],"prefix":"10.1371","volume":"21","author":[{"given":"Youjin","family":"Lee","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Seonguk","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sangil","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3384-435X","authenticated-orcid":true,"given":"Yeona","family":"Kang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"name":"for the Alzheimer\u2019s Disease Neuroimaging Initiative","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,21]]},"reference":[{"key":"pone.0341995.ref001","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/j.clinbiochem.2019.04.015","article-title":"Clinical aspects of Alzheimer\u2019s disease","volume":"72","author":"M Zv\u011b\u0159ov\u00e1","year":"2019","journal-title":"Clin Biochem"},{"issue":"9533","key":"pone.0341995.ref002","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1016\/S0140-6736(06)69113-7","article-title":"Alzheimer\u2019s disease","volume":"368","author":"K Blennow","year":"2006","journal-title":"Lancet"},{"issue":"5","key":"pone.0341995.ref003","doi-asserted-by":"crossref","first-page":"483","DOI":"10.2217\/nmt.12.55","article-title":"Activities of daily living: where do they fit in the diagnosis of Alzheimer\u2019s disease?","volume":"2","author":"GA Marshall","year":"2012","journal-title":"Neurodegener Dis Manag"},{"key":"pone.0341995.ref004","unstructured":"World Health Organization. Dementia Fact Sheet 2023. Accessed 2025 January 12. https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/dementia"},{"key":"pone.0341995.ref005","unstructured":"Centers for Disease Control and Prevention. About Alzheimer\u2019s. Accessed 2025 January 12. https:\/\/www.cdc.gov\/alzheimers-dementia\/about\/alzheimers.html"},{"issue":"2","key":"pone.0341995.ref006","doi-asserted-by":"crossref","first-page":"111","DOI":"10.31887\/DCNS.2009.11.2\/cqiu","article-title":"Epidemiology of Alzheimer\u2019s disease: occurrence, determinants, and strategies toward intervention","volume":"11","author":"C Qiu","year":"2009","journal-title":"Dialogues Clin Neurosci"},{"issue":"2","key":"pone.0341995.ref007","doi-asserted-by":"crossref","first-page":"310","DOI":"10.1111\/joim.13244","article-title":"Practical recommendations for timely, accurate diagnosis of symptomatic Alzheimer\u2019s disease (MCI and dementia) in primary care: a review and synthesis","volume":"290","author":"JL Liss","year":"2021","journal-title":"J Intern Med"},{"issue":"9518","key":"pone.0341995.ref008","doi-asserted-by":"crossref","first-page":"1262","DOI":"10.1016\/S0140-6736(06)68542-5","article-title":"Mild cognitive impairment","volume":"367","author":"S Gauthier","year":"2006","journal-title":"Lancet"},{"issue":"6","key":"pone.0341995.ref009","doi-asserted-by":"crossref","first-page":"427","DOI":"10.7326\/0003-4819-148-6-200803180-00005","article-title":"Prevalence of cognitive impairment without dementia in the United States","volume":"148","author":"BL Plassman","year":"2008","journal-title":"Ann Intern Med"},{"key":"pone.0341995.ref010","first-page":"123","article-title":"Alzheimer\u2019s Disease - Why We Need Early Diagnosis","volume":"9","author":"J Rasmussen","year":"2019","journal-title":"Degener Neurol Neuromuscul Dis"},{"issue":"4","key":"pone.0341995.ref011","doi-asserted-by":"crossref","first-page":"324","DOI":"10.2174\/156720509788929237","article-title":"Early diagnosis of Alzheimer\u2019s disease: is MCI too late?","volume":"6","author":"RC Petersen","year":"2009","journal-title":"Curr Alzheimer Res"},{"issue":"2","key":"pone.0341995.ref012","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1002\/j.2051-5545.2008.tb00159.x","article-title":"Mild cognitive impairment: searching for the prodrome of Alzheimer\u2019s disease","volume":"7","author":"PB Rosenberg","year":"2008","journal-title":"World Psychiatry"},{"key":"pone.0341995.ref013","first-page":"662","article-title":"Mild cognitive impairment and Alzheimer disease.","volume-title":"Flemming KD, editor. Mayo Clinic Neurology Board Review","author":"PW Tipton","year":"2021"},{"issue":"4","key":"pone.0341995.ref014","doi-asserted-by":"crossref","first-page":"679","DOI":"10.3233\/JAD-2011-100999","article-title":"Effects of a newly developed cognitive intervention in amnestic mild cognitive impairment and mild Alzheimer\u2019s disease: a pilot study","volume":"25","author":"VC Buschert","year":"2011","journal-title":"J Alzheimers Dis"},{"key":"pone.0341995.ref015","doi-asserted-by":"crossref","first-page":"458","DOI":"10.1093\/oso\/9780190634230.003.0019","article-title":"Management of Mild Cognitive Impairment","volume-title":"Vascular disease, alzheimer\u2019s disease, and mild cognitive impairment","author":"R Ravona-Springer","year":"2020"},{"issue":"4","key":"pone.0341995.ref016","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1038\/nrneurol.2018.9","article-title":"Imaging tau and amyloid-\u03b2 proteinopathies in Alzheimer disease and other conditions","volume":"14","author":"VL Villemagne","year":"2018","journal-title":"Nat Rev Neurol"},{"key":"pone.0341995.ref017","doi-asserted-by":"crossref","first-page":"117","DOI":"10.1016\/j.nbd.2014.05.001","article-title":"Early detection of Alzheimer\u2019s disease using PiB and FDG PET","author":"AD Cohen","year":"2014","journal-title":"Neurobiol Dis"},{"key":"pone.0341995.ref018","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1016\/j.arr.2016.02.003","article-title":"Brain fluorodeoxyglucose (FDG) PET in dementia","volume":"30","author":"T Kato","year":"2016","journal-title":"Ageing Res Rev"},{"issue":"3","key":"pone.0341995.ref019","doi-asserted-by":"crossref","first-page":"390","DOI":"10.2967\/jnumed.107.045385","article-title":"Multicenter standardized 18F-FDG PET diagnosis of mild cognitive impairment, Alzheimer\u2019s disease, and other dementias","volume":"49","author":"L Mosconi","year":"2008","journal-title":"J Nucl Med"},{"key":"pone.0341995.ref020","doi-asserted-by":"crossref","first-page":"785039","DOI":"10.1155\/2014\/246586","article-title":"A survey of FDG- and amyloid-PET imaging in dementia and GRADE analysis","volume":"2014","author":"D Perani","year":"2014","journal-title":"Biomed Res Int"},{"issue":"1","key":"pone.0341995.ref021","doi-asserted-by":"crossref","first-page":"9261","DOI":"10.1038\/s41598-020-65957-3","article-title":"Selection of the optimal intensity normalization region for FDG-PET studies of normal aging and Alzheimer\u2019s disease","volume":"10","author":"S Nugent","year":"2020","journal-title":"Sci Rep"},{"key":"pone.0341995.ref022","first-page":"118","article-title":"nnU-Net for brain tumor segmentation","volume-title":"Lecture notes in computer science","author":"F Isensee","year":"2021"},{"issue":"1","key":"pone.0341995.ref023","doi-asserted-by":"crossref","first-page":"571","DOI":"10.1007\/s10489-020-01826-w","article-title":"Deep transfer learning-based automated detection of COVID-19 from lung CT scan slices","volume":"51","author":"S Ahuja","year":"2021","journal-title":"Appl Intell (Dordr)"},{"key":"pone.0341995.ref024","doi-asserted-by":"crossref","first-page":"637386","DOI":"10.3389\/fdgth.2021.637386","article-title":"An Effective multimodal image fusion method using MRI and PET for Alzheimer\u2019s disease diagnosis","volume":"3","author":"J Song","year":"2021","journal-title":"Front Digit Health"},{"key":"pone.0341995.ref025","doi-asserted-by":"crossref","first-page":"101694","DOI":"10.1016\/j.media.2020.101694","article-title":"Convolutional neural networks for classification of Alzheimer\u2019s disease: overview and reproducible evaluation","volume":"63","author":"J Wen","year":"2020","journal-title":"Med Image Anal"},{"key":"pone.0341995.ref026","doi-asserted-by":"crossref","first-page":"1038636","DOI":"10.3389\/fncom.2023.1038636","article-title":"Conventional machine learning and deep learning in Alzheimer\u2019s disease diagnosis using neuroimaging: a review","volume":"17","author":"Z Zhao","year":"2023","journal-title":"Front Comput Neurosci"},{"issue":"1","key":"pone.0341995.ref027","doi-asserted-by":"crossref","first-page":"3404","DOI":"10.1038\/s41467-022-31037-5","article-title":"Multimodal deep learning for Alzheimer\u2019s disease dementia assessment","volume":"13","author":"S Qiu","year":"2022","journal-title":"Nat Commun"},{"issue":"1","key":"pone.0341995.ref028","doi-asserted-by":"crossref","first-page":"189","DOI":"10.1007\/s10278-022-00719-3","article-title":"An explainable convolutional neural network for the early diagnosis of alzheimer\u2019s disease from 18F-FDG PET","volume":"36","author":"LA De Santi","year":"2023","journal-title":"J Digit Imaging"},{"issue":"8","key":"pone.0341995.ref029","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1007\/s10916-022-01836-w","article-title":"Artificial intelligence on FDG PET images identifies mild cognitive impairment patients with neurodegenerative disease","volume":"46","author":"J Prats-Climent","year":"2022","journal-title":"J Med Syst"},{"issue":"7","key":"pone.0341995.ref030","first-page":"1238","article-title":"A diagnostic approach in Alzheimer\u2019s disease using three-dimensional stereotactic surface projections of fluorine-18-FDG PET","volume":"36","author":"S Minoshima","year":"1995","journal-title":"J Nucl Med"},{"issue":"4","key":"pone.0341995.ref031","doi-asserted-by":"crossref","first-page":"486","DOI":"10.1007\/s00259-005-1762-7","article-title":"Brain glucose metabolism in the early and specific diagnosis of Alzheimer\u2019s disease. FDG-PET studies in MCI and AD","volume":"32","author":"L Mosconi","year":"2005","journal-title":"Eur J Nucl Med Mol Imaging"},{"issue":"2","key":"pone.0341995.ref032","doi-asserted-by":"crossref","first-page":"632","DOI":"10.1007\/s00259-021-05603-w","article-title":"EANM procedure guidelines for brain PET imaging using [18F]FDG, version 3","volume":"49","author":"E Guedj","year":"2022","journal-title":"Eur J Nucl Med Mol Imaging"},{"issue":"1","key":"pone.0341995.ref033","doi-asserted-by":"crossref","first-page":"011013","DOI":"10.1117\/1.JMI.4.1.011013","article-title":"History and future technical innovation in positron emission tomography","volume":"4","author":"T Jones","year":"2017","journal-title":"J Med Imaging (Bellingham)"},{"key":"pone.0341995.ref034","unstructured":"Alzheimer\u2019s Disease Neuroimaging Initiative. ADNI documentation. Accessed 2025 January 12. https:\/\/adni.loni.usc.edu\/help-faqs\/adni-documentation\/"},{"key":"pone.0341995.ref035","unstructured":"Alzheimer\u2019s Disease Neuroimaging Initiative. PET 2026. Accessed 2026 May 24. https:\/\/adni.loni.usc.edu\/data-samples\/adni-data\/neuroimaging\/pet\/"},{"issue":"2","key":"pone.0341995.ref036","doi-asserted-by":"crossref","first-page":"774","DOI":"10.1016\/j.neuroimage.2012.01.021","article-title":"FreeSurfer","volume":"62","author":"B Fischl","year":"2012","journal-title":"Neuroimage"},{"issue":"3","key":"pone.0341995.ref037","doi-asserted-by":"crossref","first-page":"207","DOI":"10.3109\/14639239809001400","article-title":"A JAVA environment for medical image data analysis: initial application for brain PET quantitation","volume":"23","author":"K Mikolajczyk","year":"1998","journal-title":"Med Inform (Lond)"},{"key":"pone.0341995.ref038","unstructured":"Tan M, Le Q. EfficientNet: Rethinking model scaling for convolutional neural networks. In: Proceedings of the 36th International Conference on Machine Learning, 2019. 6105\u201314."},{"issue":"1","key":"pone.0341995.ref039","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s12065-020-00540-3","article-title":"Convolutional neural networks in medical image understanding: a survey","volume":"15","author":"DR Sarvamangala","year":"2022","journal-title":"Evol Intell"},{"issue":"12","key":"pone.0341995.ref040","doi-asserted-by":"crossref","first-page":"5017","DOI":"10.1109\/TIP.2015.2475625","article-title":"PCANet: a simple deep learning baseline for image classification?","volume":"24","author":"T-H Chan","year":"2015","journal-title":"IEEE Trans Image Process"},{"key":"pone.0341995.ref041","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pone.0341995.ref042","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International conference on knowledge discovery and data mining, 2016. 785\u201394.","DOI":"10.1145\/2939672.2939785"},{"key":"pone.0341995.ref043","article-title":"LightGBM: a highly efficient gradient boosting decision tree","volume":"30","author":"G Ke","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"pone.0341995.ref044","article-title":"CatBoost: unbiased boosting with categorical features","volume":"31","author":"L Prokhorenkova","year":"2018","journal-title":"Adv Neural Inf Process Syst"},{"issue":"1","key":"pone.0341995.ref045","doi-asserted-by":"crossref","first-page":"5979","DOI":"10.1038\/s41598-022-09954-8","article-title":"On evaluation metrics for medical applications of artificial intelligence","volume":"12","author":"SA Hicks","year":"2022","journal-title":"Sci Rep"},{"key":"pone.0341995.ref046","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1007\/978-3-319-44775-9_3","article-title":"The Mini-Mental State Examination (MMSE): update on its diagnostic accuracy and clinical utility for cognitive disorders.","volume-title":"Larner AJ, editor. Cognitive screening instruments: a practical approach","author":"AJ Mitchell","year":"2017"},{"issue":"9","key":"pone.0341995.ref047","doi-asserted-by":"crossref","first-page":"922","DOI":"10.1111\/j.1532-5415.1992.tb01992.x","article-title":"The mini-mental state examination: a comprehensive review","volume":"40","author":"TN Tombaugh","year":"1992","journal-title":"J Am Geriatr Soc"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0341995","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T17:23:50Z","timestamp":1784654630000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0341995"}},"subtitle":[],"editor":[{"given":"Emanuele","family":"Bartolini","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,21]]},"references-count":47,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,21]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0341995","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,21]]}}}