As individuals age beyond early adulthood, both physical and mental functions often experience a gradual decline. Among the most prevalent causes of significant mental deterioration in older adults are neurodegenerative diseases, which are marked by the progressive loss of neurons in the brain or peripheral nervous system.

Previous research has indicated that the mental decline and memory loss linked to neurodegenerative diseases are frequently preceded by neuropsychiatric symptoms such as depression, lack of motivation, anxiety, and irritability. However, early detection of neurodegenerative conditions based on these symptoms has been challenging.

Researchers at Chongqing Medical University and the Chongqing Key Laboratory of Oral Diseases have recently investigated a novel approach to predict cognitive decline. This method combines biological samples with machine learning to enhance early detection.

Innovative Approach to Early Detection

The study, published in Translational Psychiatry, underscores the potential of this approach for large-scale screening of older adults. It aims to identify individuals at higher risk of developing neuropsychiatric disorders or neurodegenerative diseases.

“Neuropsychiatric symptoms are early indicators of cognitive decline due to neurodegenerative diseases, and their timely detection is of the utmost importance,” stated Ping Liu, Zeng Yang, and their colleagues in their paper. “We aimed to develop and validate methods for large-scale NPS screening among elderly individuals and explore underlying metabolic mechanisms.”

Study Methodology

To conduct their research, Liu, Yang, and their team recruited 338 older adults from community health care centers in Chongqing, China. Participants provided demographic information through a questionnaire and submitted saliva samples along with samples of the bacteria in their mouths.

The researchers measured stress markers, including the hormone cortisol and cytokines, which are small proteins produced by immune cells. The collected data was divided into two datasets: one for training machine learning models and the other for validating the models’ predictive capabilities regarding neuropsychiatric symptoms.

Machine Learning Models in Action

The team developed and trained several machine learning models, including Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), and Logistic Regression (LR) models. These models were tested for their ability to identify patients at higher risk of neuropsychiatric symptoms by analyzing biomarkers from saliva and oral microbiome samples.

The XGBoost model outperformed the others, achieving the highest performance with an AUROC of 0.936 and an F1 score of 0.864. The LR model was converted into a nomogram to facilitate neuropsychiatric-risk assessment in community settings. External validation confirmed the strong predictive power, with an AUROC of 0.986 and an F1 score of 0.944.

Implications for Future Screening

The new machine learning-based screening tool developed by the researchers could soon be refined and tested in real-world clinical settings. It holds promise for helping healthcare providers detect neuropsychiatric symptoms and potentially cognitive decline early, allowing for timely therapeutic interventions and support strategies.

“The XGBoost-augmented model and nomogram offer promising tools for community-based NPS screening, while enrichment analysis provides insights into biological mechanisms,” noted Liu, Yang, and their colleagues.

Potential for Broader Applications

The initial results highlight the potential of machine learning models for analyzing biological data and detecting neuropsychiatric symptoms early. Other neuroscientists and psychiatry researchers may draw inspiration from this study to develop additional AI-based platforms for large-scale screening of older adults or other populations at higher risk of specific conditions.

This article, written by Ingrid Fadelli, edited by Gaby Clark, and fact-checked and reviewed by Robert Egan, is the result of careful human work. We rely on readers like you to keep independent science journalism alive. If this reporting matters to you, please consider a donation.

Ping Liu et al, A community screening tool for neuropsychiatric symptoms in the elderly: integrating cortisol, microbiome, and social factors with machine learning, Translational Psychiatry (2026). DOI: 10.1038/s41398-025-03797-3.

Journal information: Translational Psychiatry

© 2026 Science X Network

🔗 **Fuente:** https://medicalxpress.com/news/2026-01-aging-mental-decline-saliva-samples.html