A simple nomogram prediction model to identify relatively young patients with mild cognitive impairment who may progress to Alzheimer's disease

Aim: Construct a clinical predictive model based on easily accessible clinical features and imaging data to identify patients 65 years of age and younger with mild cognitive impairment(MCI) who may progress to Alzheimer's disease(AD). Methods: From the ADNI database, patients with MCI who were less than or equal to 65 years of age and who had been followed for 6-60 months were selected.We collected demographic data, neuropsycholog-ical test scale scores, and structural magnetic images of these patients. Clinical characteristics were then screened, and VBM and SBM analyses were performed using structural nuclear magnetic images to obtain imaging histology characteristics. Finally, predictive models were constructed combining the clinical and imaging histology characteristics. Results: The constructed nomogram has a cross-validated AUC of 0.872 in the training set and 0.867 in the verification set, and the calibration curve fits well.We also provide an online model-based forecasting tool. Conclusion: The model has good performance and uses convenience,it should be able to provide assistance in clinical work to screen relatively young MCI patients who may progress to AD. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

This study aimed to construct a clinical predictive model based on easily accessible clinical features and imaging data to identify younger patients with mild cognitive impairment (MCI) who may progress to Alzheimer's disease (AD). The model showed good performance with a high AUC and calibration curve fit, providing assistance in clinical work to screen relatively young MCI patients who may progress to AD.