A Paired Point-of-Care Ultrasound Dataset for Image Quality Enhancement and Benchmarking via a cGAN Baseline

2026-05-08

AI Summary

This study addresses the challenge of poor image quality in low-performance point-of-care ultrasound (POCUS) devices, which hinders diagnostic utility in resource-limited settings. To overcome this limitation, the authors construct the first publicly available, precisely paired dataset of POCUS and high-end ultrasound images and propose a conditional generative adversarial network (cGAN) based on the pix2pix framework. Their approach integrates a U-Net generator, a hybrid loss combining L1 and SSIM metrics, and a simulation-based pretraining strategy to substantially enhance image fidelity. Experimental results demonstrate significant improvements: the structural similarity index (SSIM) increases from 0.29 to 0.54, peak signal-to-noise ratio (PSNR) rises from 19.16 dB to 22.41 dB, and no-reference quality metrics NIQE and PIQE decrease to 4.44 and 19.99, respectively, confirming the potential of deep learning to transcend inherent hardware constraints of POCUS systems.

Abstract

Purpose: We aim to enhance the image quality of point-of-care ultrasound (POCUS) devices using deep learning and a novel paired dataset of POCUS and high-end ultrasound images. Approach: We collected the first accurately paired dataset using a custom-built automated gantry system of low-end POCUS and high-end ultrasound images. A conditional generative adversarial network (cGAN) was utilized based on the pix2pix architecture, with a U-Net generator that incorporates both L1 and structural similarity index (SSIM) losses to improve perceptual quality. Pretraining on a simulation dataset further boosts performance. Evaluation was performed on 1064 paired ex vivo tissue and phantom ultrasound image sets. Results: Our approach improves the SSIM from 0.29 to 0.54 and PSNR from 19.16 dB to 22.41 dB. No-reference metrics also indicate substantial enhancement, with the Natural Image Quality Evaluator (NIQE) and Perception-based Image Quality Evaluator (PIQE) scores dropping from 7.95 to 4.44 and 31.12 to 19.99, respectively. Conclusions: This work presents the first publicly available accurately paired dataset of low-end POCUS to high end ultrasound images. Additionally, our results demonstrate the potential of the proposed framework to overcome hardware limitations of handheld POCUS, enhancing its diagnostic value in low-resource and point-of-care settings. The POCUS-IQ Dataset is publicly available at https://github.com/NKI-MedTech-AI/POCUS-IQ.

Problem

• point-of-care ultrasound
• image quality enhancement
• paired dataset
• ultrasound imaging
• diagnostic value

Innovation

• paired ultrasound dataset
• cGAN
• image quality enhancement
• POCUS
• U-Net