Metabolic and biochemical determinants of anal disease risk in HIV-positive MSM

Background: Human immunodeficiency virus (HIV)-positive men who have sex with men (MSM) face increased risk of anal diseases, including anal intraepithelial neoplasia and carcinoma, due to immunosuppression, viral persistence, and metabolic dysfunction. Objective: This study proposes Metabolic-Biochemical Integrated Deep Risk Analytics (MBIDRA), a predictive framework for early risk stratification through multimodal feature integration. Methods: The study analysed 680 country-year observations from 170 countries, incorporating clinical, metabolic, and biochemical variables. MBIDRA integrates entropy- and mutual-information-based feature characterisation, Genetic Algorithm–Particle Swarm Optimisation (GA-PSO) feature selection, and an autoencoder-based CNN-BiLSTM-attention classifier, followed by ensemble decision fusion using gradient boosting and random forest. Results: With an 80:10:10 training-validation-test split, independent test-set evaluation achieved 96.8% accuracy and 0.97 AUC, outperforming conventional machine-learning models. Feature-importance and attention analyses identified inflammatory, metabolic, and oxidative-stress-related variables as influential predictors. Conclusion: These findings demonstrate MBIDRA’s potential for computational anal disease risk assessment in high-risk HIV-positive MSM populations.

Authors

Institutions

Publication Details

Journal
Archives of Physiology and Biochemistry
Published
2026-11-14
DOI
https://doi.org/10.1080/13813455.2026.2755757
Primary Topic
Artificial Intelligence in Healthcare
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Metabolic and biochemical determinants of anal disease risk in HIV-positive MSM

CAO Naisu, Yi Hu, Xiaqing Wang
Archives of Physiology and Biochemistry
Artificial Intelligence in Healthcare
article

Metabolic and biochemical determinants of anal disease risk in HIV-positive MSM

CAO Naisu, Yi Hu, Xiaqing Wang
article en

Abstract

Background: Human immunodeficiency virus (HIV)-positive men who have sex with men (MSM) face increased risk of anal diseases, including anal intraepithelial neoplasia and carcinoma, due to immunosuppression, viral persistence, and metabolic dysfunction. Objective: This study proposes Metabolic-Biochemical Integrated Deep Risk Analytics (MBIDRA), a predictive framework for early risk stratification through multimodal feature integration. Methods: The study analysed 680 country-year observations from 170 countries, incorporating clinical, metabolic, and biochemical variables. MBIDRA integrates entropy- and mutual-information-based feature characterisation, Genetic Algorithm–Particle Swarm Optimisation (GA-PSO) feature selection, and an autoencoder-based CNN-BiLSTM-attention classifier, followed by ensemble decision fusion using gradient boosting and random forest. Results: With an 80:10:10 training-validation-test split, independent test-set evaluation achieved 96.8% accuracy and 0.97 AUC, outperforming conventional machine-learning models. Feature-importance and attention analyses identified inflammatory, metabolic, and oxidative-stress-related variables as influential predictors. Conclusion: These findings demonstrate MBIDRA’s potential for computational anal disease risk assessment in high-risk HIV-positive MSM populations.

Archives of Physiology and Biochemistry
The Eighth Hospital of Xi'an (CN)
Artificial Intelligence in Healthcare
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon