Adaptive Few-Shot Learning PoC Ultrasound COVID-19 Diagnostic System

TL;DR

This paper introduces an adaptive few-shot learning ultrasound diagnostic system for COVID-19 that efficiently classifies disease states with limited data, demonstrating promising accuracy for point-of-care applications.

Contribution

It presents a novel adaptive FSL-based ultrasound diagnostic system with vocabulary-based feature processing for COVID-19 detection, improving efficiency and accuracy in limited-data settings.

Findings

High diagnostic accuracy on COVID-19 POCUS dataset

Efficient performance in limited-data point-of-care environments

Effective discrimination between COVID-19, pneumonia, and healthy states

Abstract

This paper presents a novel ultrasound imaging point-of-care (PoC) COVID-19 diagnostic system. The adaptive visual diagnostics utilize few-shot learning (FSL) to generate encoded disease state models that are stored and classified using a dictionary of knowns. The novel vocabulary based feature processing of the pipeline adapts the knowledge of a pretrained deep neural network to compress the ultrasound images into discrimative descriptions. The computational efficiency of the FSL approach enables high diagnostic deep learning performance in PoC settings, where training data is limited and the annotation process is not strictly controlled. The algorithm performance is evaluated on the open source COVID-19 POCUS Dataset to validate the system's ability to distinguish COVID-19, pneumonia, and healthy disease states. The results of the empirical analyses demonstrate the appropriate efficiency and accuracy for scalable PoC use. The code for this work will be made publicly available on GitHub upon acceptance.