Siyu Yang, Xintong Zhang, Xinyu Du, Peng Yan, Jing Zhang, Wei Wang, Jing Wang, Lei Zhang, Huaiqing Sun, Yin Liu, Xinran Xu, Yaxuan Di, Jin Zhong, Caiyun Wu, Jan D Reinhardt, Yu Zheng, Ting Wu
{"title":"利用深度学习预测阿尔茨海默病连续体中的认知转换。","authors":"Siyu Yang, Xintong Zhang, Xinyu Du, Peng Yan, Jing Zhang, Wei Wang, Jing Wang, Lei Zhang, Huaiqing Sun, Yin Liu, Xinran Xu, Yaxuan Di, Jin Zhong, Caiyun Wu, Jan D Reinhardt, Yu Zheng, Ting Wu","doi":"10.1186/s13195-025-01686-x","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Early diagnosis and accurate prognosis of cognitive decline in Alzheimer's disease (AD) is important to timely assignment to optimal treatment modes. We aimed to develop a deep learning model to predict cognitive conversion to guide re-assignment decisions to more intensive therapies where needed.</p><p><strong>Methods: </strong>Longitudinal data including five variable sets, i.e. demographics, medical history, neuropsychological outcomes, laboratory and neuroimaging results, from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort were analyzed. We first developed a deep learning model to predicted cognitive conversion using all five variable sets. We then gradually removed variable sets to obtained parsimonious models for four different years of forecasting after baseline within acceptable frames of reduction in overall model fit (AUC remaining > 0.8).</p><p><strong>Results: </strong>A total of 607 individuals were included at baseline, of whom 538 participants were followed up at 12 months, 482 at 24 months, 268 at 36 months and 280 at 48 months. Predictive performance was excellent with AUCs ranging from 0.87 to 0.92 when all variable sets were considered. Parsimonious prediction models that still had a good performance with AUC 0.80-0.84 were established, each only including two variable sets. Neuropsychological outcomes were included in all parsimonious models. In addition, biomarker was included at year 1 and year 2, imaging data at year 3 and demographics at year 4. Under our pre-set threshold, the rate of upgrade to more intensive therapies according to predicted cognitive conversion was always higher than according to actual cognitive conversion so as to decrease the false positive rate, indicating the proportion of patients who would have missed upgraded treatment based on prognostic models although they actually needed it.</p><p><strong>Conclusions: </strong>Neurophysiological tests combined with other indicator sets that vary along the AD continuum can improve can provide aid for clinical treatment decisions leading to improved management of the disease.</p><p><strong>Trail registration information: </strong>ClinicalTrials.gov Identifier: NCT00106899 (Registration Date: 31 March 2005).</p>","PeriodicalId":7516,"journal":{"name":"Alzheimer's Research & Therapy","volume":"17 1","pages":"41"},"PeriodicalIF":7.9000,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11823042/pdf/","citationCount":"0","resultStr":"{\"title\":\"Prediction of cognitive conversion within the Alzheimer's disease continuum using deep learning.\",\"authors\":\"Siyu Yang, Xintong Zhang, Xinyu Du, Peng Yan, Jing Zhang, Wei Wang, Jing Wang, Lei Zhang, Huaiqing Sun, Yin Liu, Xinran Xu, Yaxuan Di, Jin Zhong, Caiyun Wu, Jan D Reinhardt, Yu Zheng, Ting Wu\",\"doi\":\"10.1186/s13195-025-01686-x\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Early diagnosis and accurate prognosis of cognitive decline in Alzheimer's disease (AD) is important to timely assignment to optimal treatment modes. We aimed to develop a deep learning model to predict cognitive conversion to guide re-assignment decisions to more intensive therapies where needed.</p><p><strong>Methods: </strong>Longitudinal data including five variable sets, i.e. demographics, medical history, neuropsychological outcomes, laboratory and neuroimaging results, from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort were analyzed. We first developed a deep learning model to predicted cognitive conversion using all five variable sets. We then gradually removed variable sets to obtained parsimonious models for four different years of forecasting after baseline within acceptable frames of reduction in overall model fit (AUC remaining > 0.8).</p><p><strong>Results: </strong>A total of 607 individuals were included at baseline, of whom 538 participants were followed up at 12 months, 482 at 24 months, 268 at 36 months and 280 at 48 months. Predictive performance was excellent with AUCs ranging from 0.87 to 0.92 when all variable sets were considered. Parsimonious prediction models that still had a good performance with AUC 0.80-0.84 were established, each only including two variable sets. Neuropsychological outcomes were included in all parsimonious models. In addition, biomarker was included at year 1 and year 2, imaging data at year 3 and demographics at year 4. Under our pre-set threshold, the rate of upgrade to more intensive therapies according to predicted cognitive conversion was always higher than according to actual cognitive conversion so as to decrease the false positive rate, indicating the proportion of patients who would have missed upgraded treatment based on prognostic models although they actually needed it.</p><p><strong>Conclusions: </strong>Neurophysiological tests combined with other indicator sets that vary along the AD continuum can improve can provide aid for clinical treatment decisions leading to improved management of the disease.</p><p><strong>Trail registration information: </strong>ClinicalTrials.gov Identifier: NCT00106899 (Registration Date: 31 March 2005).</p>\",\"PeriodicalId\":7516,\"journal\":{\"name\":\"Alzheimer's Research & Therapy\",\"volume\":\"17 1\",\"pages\":\"41\"},\"PeriodicalIF\":7.9000,\"publicationDate\":\"2025-02-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11823042/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Alzheimer's Research & Therapy\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1186/s13195-025-01686-x\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CLINICAL NEUROLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Alzheimer's Research & Therapy","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s13195-025-01686-x","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}
Prediction of cognitive conversion within the Alzheimer's disease continuum using deep learning.
Background: Early diagnosis and accurate prognosis of cognitive decline in Alzheimer's disease (AD) is important to timely assignment to optimal treatment modes. We aimed to develop a deep learning model to predict cognitive conversion to guide re-assignment decisions to more intensive therapies where needed.
Methods: Longitudinal data including five variable sets, i.e. demographics, medical history, neuropsychological outcomes, laboratory and neuroimaging results, from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort were analyzed. We first developed a deep learning model to predicted cognitive conversion using all five variable sets. We then gradually removed variable sets to obtained parsimonious models for four different years of forecasting after baseline within acceptable frames of reduction in overall model fit (AUC remaining > 0.8).
Results: A total of 607 individuals were included at baseline, of whom 538 participants were followed up at 12 months, 482 at 24 months, 268 at 36 months and 280 at 48 months. Predictive performance was excellent with AUCs ranging from 0.87 to 0.92 when all variable sets were considered. Parsimonious prediction models that still had a good performance with AUC 0.80-0.84 were established, each only including two variable sets. Neuropsychological outcomes were included in all parsimonious models. In addition, biomarker was included at year 1 and year 2, imaging data at year 3 and demographics at year 4. Under our pre-set threshold, the rate of upgrade to more intensive therapies according to predicted cognitive conversion was always higher than according to actual cognitive conversion so as to decrease the false positive rate, indicating the proportion of patients who would have missed upgraded treatment based on prognostic models although they actually needed it.
Conclusions: Neurophysiological tests combined with other indicator sets that vary along the AD continuum can improve can provide aid for clinical treatment decisions leading to improved management of the disease.
期刊介绍：
Alzheimer's Research & Therapy is an international peer-reviewed journal that focuses on translational research into Alzheimer's disease and other neurodegenerative diseases. It publishes open-access basic research, clinical trials, drug discovery and development studies, and epidemiologic studies. The journal also includes reviews, viewpoints, commentaries, debates, and reports. All articles published in Alzheimer's Research & Therapy are included in several reputable databases such as CAS, Current contents, DOAJ, Embase, Journal Citation Reports/Science Edition, MEDLINE, PubMed, PubMed Central, Science Citation Index Expanded (Web of Science) and Scopus.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
