Search papers, labs, and topics across Lattice.
This study addresses the challenges of applying large vision-language models (LVLMs) to ultrasound imaging by emphasizing the importance of data scale and alignment over complex architectures. By constructing a large-scale dataset of 1.5 million ultrasound examinations with 17.7 million images and paired clinical reports, the authors demonstrate that fine-tuning a standard LVLM using low-rank adaptation (LoRA) can yield strong performance without task-specific modifications. The results reveal that this straightforward approach surpasses previous methods that relied on more intricate designs, highlighting the critical role of data organization in real-world applications.
A simple data-driven approach outperforms complex models in ultrasound understanding, revealing the power of scale and alignment.
Large vision-language models (LVLMs) have achieved strong performance across many medical imaging tasks, yet their application to ultrasound remains limited due to its inherent complexity and variability. In this work, we revisit what is truly needed to enable real-world ultrasound understanding. Instead of introducing complex architectures or elaborate training strategies, we show that data scale and clinically faithful data alignment are the key factors. We construct a large-scale dataset of 1.5M real-world ultrasound examinations, containing 17.7M images, multi-organ coverage, and paired uncurated clinical reports. Crucially, we organize the data at the examination level, aligning multiple images with their corresponding reports to reflect real clinical workflows. We then fine-tune a standard LVLM using low-rank adaptation (LoRA) on this dataset without task-specific modifications. Surprisingly, this simple recipe already leads to strong performance across diverse ultrasound understanding tasks, outperforming prior methods designed with more complex pipelines. Beyond these results, we present model and data scaling analyses that provide insights into the role of scale in ultrasound LVLMs.
