USFM is the first foundation model for medical ultrasound images, developed and maintained by the Laboratory of Medical Imaging and Artificial Intelligence, Fudan University. USFM aims to accelerate the modeling of existing medical ultrasound image analysis tasks with high performance and efficiency (less labeled data and fewer training epochs).

在本文中，我们提出了一种通用的 US 基础模型（名为 USFM），该模型适用于不同的任务和器官，以实现高效的 US 图像分析。 数据库不足： 充足且多样化的 US 图像数据库是建立全面 US 基础模型的基础。 由于数据采集的复杂性和患者隐私的问题，从多种来源收集涵盖多种器官的各种成像设备的 US 图像非常困难。 此外，发病率较高的器官在数据库中占主导地位，导致不同器官比例失衡。 低质量： US 图像通常存在固有的质量问题，如分辨率、对比度和信噪比较低，导致有 …

2023年12月30日 · In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale Multi-organ, Multi-center, and Multi-device US database was built, comprehensively containing over two million US images.

2024年8月1日 · In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale M ulti-organ, M ulti-center, and M ulti-device US database was built, comprehensively containing over two million US images.

2024年8月1日 · In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale M ulti-organ, M ulti-center, and Multi-device US database was built, comprehensively containing over two million US images.

USFM is the first foundation model for medical ultrasound images, developed and maintained by the Laboratory of Medical Imaging and Artificial Intelligence, Fudan University. USFM aims to accelerate the modeling of existing medical ultrasound image analysis tasks with high performance and efficiency (less labeled data and fewer training epochs).

World's first open-source platform for medical foundation models. 10+ medical data modalities targeting a variety of clinical and research problems. Pioneering works of the new learning paradigm using foundation models, including pre-trained models, code, and data.

In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale ulti-organ, ulti-center, and ulti-device US database was built, comprehensively containing over two …

2023年12月30日 · In this paper, we present a universal US foundation model, named USFM, generalized to diverse tasks and organs towards label efficient US image analysis. First, a large-scale Multi-organ, Multi-center, and Multi-device US database was built, comprehensively containing over two million US images.

USFM has attempted to “unify” a long history of SFM type scripture markup “standards”, some of which were more or less strict in their tolerance for format-oriented markers. The primary focus in USFM development was on unification, not markup creation.