Pan-infection Foundation Framework Enables Multiple Pathogen Prediction

2024-12-31

AI Summary

Current infection diagnostic models suffer from limited sample sizes and coarse pathogen categorization, hindering accurate discrimination between bacterial and viral infections or identification of specific pathogens. To address this, we construct the first large-scale pan-infectious host transcriptomic dataset—comprising 11,247 samples across 89 independent cohorts—and propose a novel “pan-infectious foundation model + knowledge distillation” paradigm. First, we train a high-accuracy foundational diagnostic model (AUC = 0.97); then, via knowledge distillation, we derive lightweight, pathogen-specific models achieving exceptional performance: *Staphylococcus aureus* (AUC = 0.99), *Streptococcus* spp. (0.94), HIV (0.93), RSV (0.94), and sepsis (0.99). This framework enables, for the first time, transferable representation learning across diverse pathogens and clinical conditions (e.g., from pan-infectious states to sepsis), supporting clinically deployable, high-accuracy, resource-efficient diagnostics.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Machine Learning: Large Multimodal Models (LMMs)
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Host-response-based diagnostics can improve the accuracy of diagnosing bacterial and viral infections, thereby reducing inappropriate antibiotic prescriptions. However, the existing cohorts with limited sample size and coarse infections types are unable to support the exploration of an accurate and generalizable diagnostic model. Here, we curate the largest infection host-response transcriptome data, including 11,247 samples across 89 blood transcriptome datasets from 13 countries and 21 platforms. We build a diagnostic model for pathogen prediction starting from a pan-infection model as foundation (AUC = 0.97) based on the pan-infection dataset. Then, we utilize knowledge distillation to efficiently transfer the insights from this"teacher"model to four lightweight pathogen"student"models, i.e., staphylococcal infection (AUC = 0.99), streptococcal infection (AUC = 0.94), HIV infection (AUC = 0.93), and RSV infection (AUC = 0.94), as well as a sepsis"student"model (AUC = 0.99). The proposed knowledge distillation framework not only facilitates the diagnosis of pathogens using pan-infection data, but also enables an across-disease study from pan-infection to sepsis. Moreover, the framework enables high-degree lightweight design of diagnostic models, which is expected to be adaptively deployed in clinical settings.

Problem

• Infectious Disease Diagnosis
• Bacterial vs Viral Infections
• Specific Pathogen Identification

Innovation

• Knowledge Distillation
• Pathogen Prediction Model
• Clinical Application Enhancement