Highly accurate and rapid, host-based tools for differentiating between bacterial and viral infection have great potential to support appropriate antibiotic use, reduce clinical uncertainty and care variance, and improve operational efficiencies. The challenge is to identify best practices for implementation to ensure these added value dimensions are realized in real-world settings. This talk will consider the interplay of clinical workflows and target patient populations in clinical evidence approaches to demonstrate that these new adjunctive tests improve patient outcomes and management.

The diagnosis and prognosis of patients with suspected acute infection and suspected sepsis remains a clinical challenge. Novel tests offer the potential to improve patient and health system outcomes. We will discuss how these new technologies work and how they might be clinically applied.

This presentation will focus on FebriDx, a novel technology designed to detect and rule out bacterial acute respiratory infection at the point-of-care. FebriDx uses two host response biomarkers to aid clinicians in balancing their need to rapidly detect and treat bacterial infection while also contributing to antimicrobial stewardship efforts by accurately ruling out bacterial infection and avoiding the prescription of unnecessary antibiotics. This presentation will review the science behind the test and the evidence that supports clinical utilization.

Tuberculosis case finding is hampered by the need to collect sputum. Many patients cannot produce a testable sputum samples, and the process is logistically challenging for all patients in community settings. We have found that most pulmonary tuberculosis cases can instead be identified by testing tongue swab samples, which are easy to collect from any person, in any setting. In the same way that non-invasive nasal swabbing replaced nasopharyngeal swabs and enabled mass screening for COVID-19, tongue swabbing could enable entirely new strategies for tuberculosis diagnosis, screening, and prevention of transmission. 

Recent studies on cough-based respiratory disease screening suggest promising performance of cough classifiers but potential biases in model training and dataset quality preclude robust conclusions. We enrolled subjects with pulmonary TB (N=149) and controls with other respiratory illnesses (N=46) in Nairobi, Kenya.  We collected a two-hour audio recording per subject to build a dataset with 33K passive coughs. We trained and evaluated the dataset with a ResNet18-based machine learning cough classifier called TBScreen. Passive cough spectral features distinguished TB and non-TB groups. TBScreen can be deployed on a smartphone and identifies cough features associated with bacterial burden and disease severity.

• NIH-wide SBIR and STTR
• Funding opportunities
• Application tips
• Assistance and initiatives for awardees and applicants​

NGS-based diagnostics for infectious diseases are advancing from pioneering studies to practice implementation. This talk will provide an overview of currently available assays from reference laboratories, as well as perspectives on independent assay development. Topics will include best-use practices to optimize value, the importance of collaborative diagnostic stewardship, and a focus on data quality for patient safety.

Next-generation sequencing (NGS) has found multiple applications in infectious diseases diagnostics, including infection prevention, pathogen detection, and antimicrobial resistance prediction. While great advancement has been made in these areas, we must identify and address remaining challenges to move infectious diseases NGS into routine practice.

As new applications of infectious NGS testing are being implemented in both clinical and commercial laboratories, the medical microbiologist plays a key role in assay interpretation and communication of results. Here we will utilize several case-based examples of the challenges in NGS result reporting and suggestions for best practice.

Next-generation sequencing methods allow for comprehensive detection of antimicrobial resistance genes. Here, we will discuss the current NGS technologies and approaches to study antimicrobial resistance and predict antimicrobial susceptibility.

Invasive Fungal Diseases are a significant cause of mortality and morbidity, especially in immunocompromised patients population. Metagenomics is being explored as an emerging method for the diagnosis of invasive fungal diseases but is still not widely available, costly and its performance characteristic remains to be established. This talk will discuss available molecular diagnostic methods that are currently more readily available, cost-effective and that may be used to improve the diagnosis of fungal diseases

I will describe UCLA Health's laboratory stewardship program and governance structure as an example of how to get started with effective laboratory stewardship for a large health system. I will discuss our program's initial successes, including establishing oversight and data monitoring for referral lab (send-out) testing and inpatient genetic testing.

Increasingly complex and expensive sequencing-based tests for the diagnosis of infectious diseases are becoming more commonly used. They can be extremely helpful in some instances, and highly misleading or wasteful in others. The literature is variable and influenced by publication bias. This case-based presentation highlights challenges in stewardship and interpretation of plasma metagenomic sequencing from a laboratory perspective.