Abstract:Super-resolution ultrasound imaging with ultrasound localization microscopy (ULM) offers a high-resolution view of microvascular structures. Yet, ULM image quality heavily relies on precise microbubble (MB) detection. Despite the crucial role of localization algorithms, there has been limited focus on the practical pitfalls in MB detection tasks such as setting the detection threshold. This study examines how False Positives (FPs) and False Negatives (FNs) affect ULM image quality by systematically adding controlled detection errors to simulated data. Results indicate that while both FP and FN rates impact Peak Signal-to-Noise Ratio (PSNR) similarly, increasing FP rates from 0\% to 20\% decreases Structural Similarity Index (SSIM) by 7\%, whereas same FN rates cause a greater drop of around 45\%. Moreover, dense MB regions are more resilient to detection errors, while sparse regions show high sensitivity, showcasing the need for robust MB detection frameworks to enhance super-resolution imaging.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to balance the influence of false positives (FP) and false negatives (FN) on image quality when detecting microbubbles (MB) in ultrasound localization microscopy (ULM) in super - resolution ultrasound imaging. Specifically, the research aims to: 1. **Evaluate the detection threshold**: Explore the specific impacts of different FP and FN rates on the quality of super - resolution ultrasound images, especially how these errors affect the peak signal - to - noise ratio (PSNR) and the structural similarity index (SSIM). 2. **Analyze the sensitivity of different regions**: By introducing controlled detection errors into the simulated data, study the differences in the sensitivity of dense and sparse microbubble regions to FP and FN. 3. **Optimize the detection framework**: Provide suggestions on how to set a more robust microbubble detection framework to improve the quality of super - resolution imaging. ### Research Background Super - resolution ultrasound imaging (SR - US) combined with ultrasound localization microscopy (ULM) can observe microvascular structures with high resolution. However, the ULM image quality highly depends on the accurate detection of microbubbles (MB). Although the localization algorithm is crucial, there are still many challenges in practical applications, such as tasks like setting the detection threshold. These problems may lead to the occurrence of false positives and false negatives, which in turn affect the final image quality. ### Main Findings - **The influence of FP and FN on image quality**: - In all regions, when the FP rate increases from 0% to 20%, the SSIM decreases by about 7%, while the same FN rate causes the SSIM to decrease by about 45%. - Dense microbubble regions are more robust to detection errors, while sparse regions show higher sensitivity. - **The influence of different frequencies**: - Data with a high central frequency is more sensitive to FN, especially in sparse regions. Adding 10% of FN will cause the SSIM to decrease by about 10%. ### Conclusion In order to optimize the quality of super - resolution ultrasound images, it is necessary to adjust the detection parameters according to the microbubble density and the imaging frequency. In particular, in sparse regions, the FP and FN rates should be strictly controlled to prevent the rapid decrease of PSNR and ensure high signal fidelity. In addition, although using a higher frequency can improve the image quality, special attention should also be paid to the control of the FN rate to avoid a significant decrease in the image quality. Through these research results, the author provides practical guidance for clinicians and researchers to optimize the ULM image quality and adjust the detection sensitivity according to the microbubble density and the imaging frequency.

Evaluating Detection Thresholds: The Impact of False Positives and Negatives on Super-Resolution Ultrasound Localization Microscopy

Microbubble Identification Based on Decision Theory for Ultrasound Localization Microscopy

Understanding the effects of microbubble concentration on localization accuracy in super-resolution ultrasound imaging

Ensemble Learning for Microbubble Localization in Super-Resolution Ultrasound

Overlapping Microbubble Localization Based on Multiscale Statistical Features for Ultrasound Super-Resolution Imaging

Improving sub-pixel accuracy in ultrasound localization microscopy using supervised and self-supervised deep learning

Analytic Optimization-Based Microbubble Tracking in Ultrasound Super-Resolution Microscopy

Learning Super-Resolution Ultrasound Localization Microscopy from Radio-Frequency Data

Ultrasound Localization Microscopy and Super-Resolution: A State of the Art

Deep learning for fast super-resolution ultrasound microvessel imaging

A Review of Clinical Applications for Super-resolution Ultrasound Localization Microscopy

Deep Learning for Ultrasound Localization Microscopy

Frame rate effects and their compensation on super-resolution microvessel imaging using ultrasound localization microscopy

Qulm-Dl: Quantitative Ultrasound Localization Microscopy Via Deep Learning

Super-Resolved Microbubble Localization in Single-Channel Ultrasound RF Signals Using Deep Learning

Principles and techniques of microbubble localization and tracking in super-resolution ultrasound microvascular imaging

Doppler Slicing for Ultrasound Super-Resolution Without Contrast Agents

[Super-resolution ultrasound imaging : Methods and applications]

Deep Learning in Ultrasound Localization Microscopy: Applications and Perspectives

Enhancing super-resolution ultrasound localisation through multi-frame deconvolution exploiting spatiotemporal coherence

Kalman Filter-Based Microbubble Tracking for Robust Super-Resolution Ultrasound Microvessel Imaging