CREATE PhD fellow Kondwelani Mateyo will be presenting his research on 'Artificial-Intelligence-Assisted Lung Ultrasound in Pulmonary Tuberculosis: Evaluating Image Acquisition Protocols and Probe Accuracy, and Factors influencing Acceptance and Adoption in Zambia' at his upgrading seminar on the 8th of April 2024, at 14:00 BST.
Background:
Tuberculosis (TB) remains a significant global health challenge, ranking as the second leading infectious cause of mortality worldwide. Efficient screening pathways and timely treatment of new cases are crucial for controlling TB spread and minimizing missed diagnoses. However, current diagnostic tools are often costly and require specialized expertise. Point of care (POC) lung ultrasound (LUS), when combined with artificial intelligence (AI), has the potential to address these challenges by reducing the need for human interpretation and specialized operators.
Previous studies have demonstrated the utility of LUS in TB diagnosis. However, questions remain regarding the optimal LUS protocol, probe type, and the development of AI-based image interpretation.
In healthcare settings where non-technical staff manage facilities, adopting a simplified lung ultrasound (LUS) protocol would be advantageous if it meets the World Health Organization’s (WHO) criteria for an effective screening tool.
Aim:
This PhD research aims to assess the accuracy of a simplified POC LUS protocol compared to a more comprehensive one in pulmonary TB (PTB) screening. Additionally, it seeks to determine the more accurate of two probe for LUS image acquisition and explore the acceptability, enablers, and barriers to AI-assisted LUS (AI-LUS) in PTB screening.
Methods:
The research methodology will comprise a systematic review and a convergent parallel mixed methods study. The systematic review will evaluate existing literature on various LUS image acquisition protocols and probes for diagnosing PTB. The mixed methods approach will encompass a randomized controlled trial, participant survey, and qualitative research approach. Presumptive PTB patients presenting to Kanyama Hospital in Lusaka, Zambia, will be randomized to either a simplified or comprehensive LUS protocol and assigned either a curvilinear or linear LUS probe. Surveys, focused group discussions (FGDs), and in-depth interviews (IDIs) will explore LUS acceptability and identify enablers and barriers to PTB screening.
If you would like to join the seminar, please use the details below :
Zoom - Meeting ID: 942 2287 1476, Password: 763184