Dai Haoran, Xiao Yuyao, Fu Caixia, Grimm Robert, von Busch Heinrich, Stieltjes Bram, Choi Moon Hyung, Xu Zhoubing, Chabin Guillaume, Yang Chun, Zeng Mengsu
Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai, China.
MR Application Development, Siemens Shenzhen Magnetic Resonance Ltd., Shenzhen, China.
J Magn Reson Imaging. 2025 Jan;61(1):111-120. doi: 10.1002/jmri.29404. Epub 2024 Jun 3.
The number of focal liver lesions (FLLs) detected by imaging has increased worldwide, highlighting the need to develop a robust, objective system for automatically detecting FLLs.
To assess the performance of the deep learning-based artificial intelligence (AI) software in identifying and measuring lesions on contrast-enhanced magnetic resonance imaging (MRI) images in patients with FLLs.
Retrospective.
395 patients with 1149 FLLs.
FIELD STRENGTH/SEQUENCE: The 1.5 T and 3 T scanners, including T1-, T2-, diffusion-weighted imaging, in/out-phase imaging, and dynamic contrast-enhanced imaging.
The diagnostic performance of AI, radiologist, and their combination was compared. Using 20 mm as the cut-off value, the lesions were divided into two groups, and then divided into four subgroups: <10, 10-20, 20-40, and ≥40 mm, to evaluate the sensitivity of radiologists and AI in the detection of lesions of different sizes. We compared the pathologic sizes of 122 surgically resected lesions with measurements obtained using AI and those made by radiologists.
McNemar test, Bland-Altman analyses, Friedman test, Pearson's chi-squared test, Fisher's exact test, Dice coefficient, and intraclass correlation coefficients. A P-value <0.05 was considered statistically significant.
The average Dice coefficient of AI in segmentation of liver lesions was 0.62. The combination of AI and radiologist outperformed the radiologist alone, with a significantly higher detection rate (0.894 vs. 0.825) and sensitivity (0.883 vs. 0.806). The AI showed significantly sensitivity than radiologists in detecting all lesions <20 mm (0.848 vs. 0.788). Both AI and radiologists achieved excellent detection performance for lesions ≥20 mm (0.867 vs. 0.881, P = 0.671). A remarkable agreement existed in the average tumor sizes among the three measurements (P = 0.174).
AI software based on deep learning exhibited practical value in automatically identifying and measuring liver lesions.
Stage 2.
全球范围内，通过影像学检查发现的肝脏局灶性病变（FLLs）数量有所增加，这凸显了开发一种强大、客观的自动检测FLLs系统的必要性。
评估基于深度学习的人工智能（AI）软件在识别和测量FLLs患者的对比增强磁共振成像（MRI）图像上病变的性能。
回顾性研究。
395例患者，共1149个FLLs。
场强/序列：1.5T和3T扫描仪，包括T1加权、T2加权、扩散加权成像、同/反相位成像以及动态对比增强成像。
比较AI、放射科医生及其联合诊断的性能。以20mm作为截断值，将病变分为两组，然后进一步分为四个亚组：<10mm、10 - 20mm、20 - 40mm和≥40mm，以评估放射科医生和AI检测不同大小病变的敏感性。我们比较了122个手术切除病变的病理大小与使用AI测量的大小以及放射科医生测量的大小。
McNemar检验、Bland - Altman分析、Friedman检验、Pearson卡方检验、Fisher精确检验、Dice系数和组内相关系数。P值<0.05被认为具有统计学意义。
AI在肝脏病变分割中的平均Dice系数为0.62。AI与放射科医生联合诊断的表现优于放射科医生单独诊断，检测率（0.894对0.825）和敏感性（0.883对0.806）显著更高。在检测所有<20mm的病变时，AI的敏感性显著高于放射科医生（0.848对0.788）。对于≥20mm的病变，AI和放射科医生均具有出色的检测性能（0.867对0.881，P = 0.671）。三种测量方法在平均肿瘤大小方面存在显著一致性（P = 0.174）。
基于深度学习的AI软件在自动识别和测量肝脏病变方面具有实用价值。
2级。
