CoChat

The DOI matches its Crossref record; 57 of the 58 references it cites were found in a registry or an index; 15 of the 26 sources its claims cite were checked for support.

Integrated algorithm combining plasma biomarkers and cognitive assessments accurately predicts brain β-amyloid pathology

Open access

Abstract

BACKGROUND: Accurate prediction of cerebral amyloidosis with easily available indicators is urgently needed for diagnosis and treatment of Alzheimer's disease (AD). METHODS: We examined plasma Aβ42, Aβ40, T-tau, P-tau181, and NfL, with APOE genotypes, cognitive test scores and key demographics in a large Chinese cohort (N = 609, aged 40 to 84 years) covering full AD spectrum. Data-driven integrated computational models were developed to predict brain β-amyloid (Aβ) pathology. RESULTS: Our computational models accurately predict brain Aβ positivity (area under the ROC curves (AUC) = 0.94). The results are validated in Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. Particularly, the models have the highest prediction power (AUC = 0.97) in mild cognitive impairment (MCI) participants. Three levels of models are designed with different accuracies and complexities. The model which only consists of plasma biomarkers can predict Aβ positivity in amnestic MCI (aMCI) patients with AUC = 0.89. Generally the models perform better in participants without comorbidities or family histories. CONCLUSIONS: The innovative integrated models provide opportunity to assess Aβ pathology in a non-invasive and cost-effective way, which might facilitate AD-drug development, early screening, clinical diagnosis and prognosis evaluation.

Overview

CoChat Summary

## TL;DR

This study built data-driven decision-tree models combining plasma biomarkers, APOE genotype, cognitive tests, and demographics to predict brain β-amyloid (Aβ) pathology in a Chinese cohort of 609 participants. The best performance was in MCI, especially aMCI, and the models were validated in an independent ADNI cohort.

## Key findings

  • In the full Chinese cohort (N = 609), the integrated model predicted brain Aβ positivity with an AUC of 0.94; the simplified “best model” had an AUC of 0.83, and the 3-step refined model had an AUC of 0.71.
  • Performance was highest in the MCI subgroup, with an AUC of 0.93 for the best model; subgroup AUCs were 0.82 for CN, 0.82 for SCD, and 0.86 for dementia.
  • In aMCI, the model performed especially well: AUCs were 0.97 (full), 0.96 (best), and 0.89 (refined); the refined aMCI model used only plasma P-tau181, Aβ40, and Aβ42.
  • In the ADNI validation cohort (284 cases), performance remained similar: AUCs of 0.96 (full), 0.88 (best), and 0.75 (refined) in the whole dataset, with 0.95/0.93/0.87 in MCI.

## Methods

  • Consecutive participants were enrolled from Shanghai Sixth People’s Hospital; 609 individuals aged 40 to 84/85 years with blood tests, MRI, and 18F-florbetapir PET within 3 months of blood sampling were included.
  • The study measured plasma Aβ42, Aβ40, T-tau, P-tau181, and NfL, plus APOE genotypes, cognitive screening scores (MMSE, MoCA-BC, ACE-III-CV) and neuropsychological tests; diagnoses included CN, SCD, MCI (including aMCI), Alzheimer’s disease dementia, and non-Alzheimer’s disease dementia.
  • The main analysis used data-driven recursive partitioning/decision tree models with cross-validation, model simplification by variable reduction, and AUC/ROC evaluation; models were also validated in the ADNI cohort (284 cases).

## Limitations

  • The paper explicitly states that, although the sample was the largest in China to their knowledge, it was smaller than some American and European studies such as ADNI and BioFINDER.
  • Evident from the text: the Chinese cohort was from one hospital in Shanghai, so generalizability beyond this setting may be limited.
  • Evident from the text: not all participants completed all neuropsychological assessments, which the authors note when justifying decision-tree methods’ tolerance for missing values.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s43856-023-00295-9

Cites

57 of 57 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Integrated models in the whole dataset. Model selection process and performance for predicting PET positivity in all the participants (n = 609). a The decision tree model selection process. The full model was generated with all covariates. The best model was pruned with the best cross-validation error rate and deleted as many variables as possible while maintaining similar performance. The refined model restricted the tree depths to three. b Receiver operating characteristic (ROC) curves of the different models for discriminating Aβ-PET positive versus negative.
Fig. 1 Integrated models in the whole dataset. Model selection process and performance for predicting PET positivity in all the participants (n = 609). a The decision tree model selection process. The full model was generated with all covariates. The best model was pruned with the best cross-validation error rate and deleted as many variables as possible while maintaining similar performance. The refined model restricted the tree depths to three. b Receiver operating characteristic (ROC) curves of the different models for discriminating Aβ-PET positive versus negative.CC BY · PubMed Central
Fig. 2 Integrated models in individual groups. Area under the Receiver Operating Characteristic (AUROC) values and the corresponding 95% confidence intervals of the established models for predicting Aβ-PET positivity in different diagnostic groups (All participants: n = 609; CN: n = 238; SCD: n = 118; MCI: n = 135; Dementia: n = 118) were shown.
Fig. 2 Integrated models in individual groups. Area under the Receiver Operating Characteristic (AUROC) values and the corresponding 95% confidence intervals of the established models for predicting Aβ-PET positivity in different diagnostic groups (All participants: n = 609; CN: n = 238; SCD: n = 118; MCI: n = 135; Dementia: n = 118) were shown.CC BY · PubMed Central
Fig. 3 Tree view of aMCI refined model. Decision Tree diagram of the refined model in the aMCI group (n = 93). The tree started from the root node. If the value met the current condition, the workflow would go to the left node; otherwise, the right node. The green nodes indicated PET negative and the red nodes indicated PET positive. Note that all values were transformed to z-score prior to fitting the model. Cut-off values were shown in z-score values.
Fig. 3 Tree view of aMCI refined model. Decision Tree diagram of the refined model in the aMCI group (n = 93). The tree started from the root node. If the value met the current condition, the workflow would go to the left node; otherwise, the right node. The green nodes indicated PET negative and the red nodes indicated PET positive. Note that all values were transformed to z-score prior to fitting the model. Cut-off values were shown in z-score values.CC BY · PubMed Central
Fig. 4 Model validation in ADNI cohort. AUROC values and the corresponding 95% confidence intervals of the established model for predicting Aβ-PET positivity in different diagnostic groups and in different cohorts. The n number of each group is showed as follows: All participants: Training with SCD & non-AD: n = 609; Training without SCD & non-AD: n = 462; ADNI: n = 284. CN: Training: n = 238; ADNI: n = 97. MCI: Training: n = 135; ADNI: n = 124.
Fig. 4 Model validation in ADNI cohort. AUROC values and the corresponding 95% confidence intervals of the established model for predicting Aβ-PET positivity in different diagnostic groups and in different cohorts. The n number of each group is showed as follows: All participants: Training with SCD & non-AD: n = 609; Training without SCD & non-AD: n = 462; ADNI: n = 284. CN: Training: n = 238; ADNI: n = 97. MCI: Training: n = 135; ADNI: n = 124.CC BY · PubMed Central
Fig. 5 Prediction of disease stages by plasma biomarkers alone. ROC plots for showing the efficiency of the integrated model for distinguishing the patient with different dementia statuses. (CN: n = 238; SCD: n = 118; MCI: n = 135; Alzheimer’s disease: n = 89).
Fig. 5 Prediction of disease stages by plasma biomarkers alone. ROC plots for showing the efficiency of the integrated model for distinguishing the patient with different dementia statuses. (CN: n = 238; SCD: n = 118; MCI: n = 135; Alzheimer’s disease: n = 89).CC BY · PubMed Central

Verification

Claims the paper makes

6 Valid · 1 Not valid · 8 Not checked · 36 Own results and interpretations

24 more claims

Create a free account to see all 51 claims and to verify the 2 references still open.

See all claims — free account
  • Probably not a substantial claim. 14
  • It may state two separate claims as one. 9
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Citations

Cites

Show all 57 references

Cited by

See all 27

Metadata

Authors
Pan, F., Huang, Y., Cai, X., Wang, Y., Guan, Y., Deng, J., Yang, D., Zhu, J., Zhao, Y., Xie, F., Fang, Z., & Guo, Q.
Affiliations
  • Shanghai Jiao Tong University (CN)
  • Fudan University (CN)
Corresponding author
  • Xie, Fang
  • Fang, Zhuo
  • Guo, Qihao
Venue
Communications Medicine
Published
2023-05-10
Volume
3
Issue
1
Pages
65-65
ISSN
2730-664X (ISSN-L)
OpenAlex
W4376113744
Cited by
27
References
58
Citations by year
2026: 11 · 2025: 8 · 2024: 6 · 2023: 2
FWCI
5.47
Type
journal article
Language
en
Open access
Licence
Locations
  • Communications Medicine (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
National Natural Science Foundation of China (2018B030336001, 82171198, 2018SHZDZX01)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Alzheimer's disease research and treatments
  • Health Sciences › Medicine › Oncology › Cancer-related cognitive impairment studies
Keywords
Alzheimer's disease, cerebral amyloidosis, plasma biomarkers, amyloid-beta 42, Aβ40, tau protein, phosphorylated tau, neurofilament light chain, APOE genotype, cognitive assessment, mild cognitive impairment, amnestic MCI, prediction model, early diagnosis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Integrated algorithm combining plasma biomarkers and cognitive assessments accurately predicts brain β-amyloid pathology. https://app.cochat.ai/papers/10.1038/s43856-023-00295-9
The paper
Pan, F., Huang, Y., Cai, X., Wang, Y., Guan, Y., Deng, J., Yang, D., Zhu, J., Zhao, Y., Xie, F., Fang, Z., & Guo, Q. (2023). Integrated algorithm combining plasma biomarkers and cognitive assessments accurately predicts brain β-amyloid pathology. Communications Medicine. https://doi.org/10.1038/s43856-023-00295-9

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.