Ding Wenzhen, Meng Yaqing, Ma Jun, Pang Chuan, Wu Jiapeng, Tian Jie, Yu Jie, Liang Ping, Wang Kun
Department of Interventional Ultrasound, Chinese PLA General Hospital, Beijing 100853, China.
CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, Beijing, China; School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China.
J Hepatol. 2025 Jan 21. doi: 10.1016/j.jhep.2025.01.011.
BACKGROUND & AIMS: Accurate multi-classification is a prerequisite for appropriate management of focal liver lesions (FLLs). Ultrasound is the most common imaging examination but lacks accuracy. Contrast-enhanced ultrasound (CEUS) offers better performance but is highly dependent on operator experience. Therefore, we aimed to develop a CEUS-based artificial intelligence (AI) model for FLL multi-classification and evaluate its performance in multicenter clinical tests.
Since January 2017 to December 2023, CEUS videos, immunohistochemical biomarkers and clinical information on solid FLLs >1 cm in adults were collected from 52 centers to build and test the model. The model was developed to classify FLLs into six types: hepatocellular carcinoma, hepatic metastasis, intrahepatic cholangiocarcinoma, hepatic hemangioma, hepatic abscess and others. First, Module-Disease, Module-Biomarker and Module-Clinic were built in training set A and a validation set. Then, three modules were aggregated as Model in training set B and an internal test set. Model performance was compared with CEUS and MRI radiologists in three external test sets.
In total 3,725 FLLs from 52 centers were divided into training set A (n = 2,088), the validation set (n = 592), training set B (n = 234), the internal test set (n = 110), and external test sets A (n = 113), B (n = 276) and C (n = 312). In external test sets A, B and C, Model achieved significantly better performance (accuracy from 0.85 to 0.86) than junior CEUS radiologists (0.59-0.73), and comparable performance to senior CEUS radiologists (0.79-0.85) and senior MRI radiologists (0.82-0.86). In multiple subgroup analyses on demographic characteristics, tumor characteristics and ultrasound devices, its accuracy ranged from 0.79 to 0.92.
CEUS-based Model provides accurate multi-classification of FLLs. It holds promise for a wide range of populations, especially those in remote areas who have difficulty accessing MRI.
NCT04682886.
Ultrasound is the most common imaging examination for screening focal liver lesions (FLLs), but it lacks accuracy for multi-classification, which is a prerequisite for appropriate clinical management. Contrast-enhanced ultrasound (CEUS) offers better diagnostic performance but relies on the experience of radiologists. We developed a CEUS-based model (Model) that can help junior CEUS radiologists to achieve comparable diagnostic ability as senior CEUS radiologists and senior MRI radiologists. The combination of an ultrasound device, CEUS examination and Model means that even patients in remote areas can be accurately diagnosed through examination by junior radiologists.
准确的多分类是局灶性肝病变（FLLs）合理管理的前提条件。超声是最常见的影像学检查方法，但缺乏准确性。超声造影（CEUS）表现更佳，但高度依赖操作者经验。因此，我们旨在开发一种基于CEUS的人工智能（AI）模型用于FLLs的多分类，并在多中心临床试验中评估其性能。
自2017年1月至2023年12月，从52个中心收集了成人实性FLLs大于1 cm的CEUS视频、免疫组化生物标志物和临床信息，用于构建和测试模型。该模型旨在将FLLs分为六种类型：肝细胞癌、肝转移瘤、肝内胆管癌、肝血管瘤、肝脓肿及其他。首先，在训练集A和验证集中构建疾病模块、生物标志物模块和临床模块。然后，在训练集B和内部测试集中将三个模块整合为模型。在三个外部测试集中将模型性能与CEUS和MRI放射科医生进行比较。
来自52个中心的3725个FLLs被分为训练集A（n = 2088）、验证集（n = 592）、训练集B（n = 234）、内部测试集（n = 110）以及外部测试集A（n = 113）、B（n = 276）和C（n = 312）。在外部测试集A、B和C中，模型的表现（准确率为0.85至0.86）显著优于初级CEUS放射科医生（0.59 - 0.73），与高级CEUS放射科医生（0.79 - 0.85）和高级MRI放射科医生（0.82 - 0.86）相当。在关于人口统计学特征、肿瘤特征和超声设备的多项亚组分析中，其准确率范围为0.79至0.92。
基于CEUS的模型可提供准确的FLLs多分类。它对广泛人群具有应用前景，尤其是那些难以进行MRI检查的偏远地区人群。
NCT04682886。
超声是筛查局灶性肝病变（FLLs）最常用的影像学检查方法，但多分类缺乏准确性，而这是合理临床管理的前提条件。超声造影（CEUS）具有更好的诊断性能，但依赖放射科医生的经验。我们开发了一种基于CEUS的模型（Model），可帮助初级CEUS放射科医生达到与高级CEUS放射科医生和高级MRI放射科医生相当的诊断能力。超声设备、CEUS检查和模型相结合意味着即使是偏远地区的患者也可通过初级放射科医生的检查得到准确诊断。
