Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
Center of Excellence for Innovation and Endoscopy in Gastrointestinal Oncology, Division of Gastroenterology, Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
Sci Rep. 2022 May 11;12(1):7749. doi: 10.1038/s41598-022-11506-z.
Despite the wide availability of ultrasound machines for hepatocellular carcinoma surveillance, an inadequate number of expert radiologists performing ultrasounds in remote areas remains a primary barrier for surveillance. We demonstrated feasibility of artificial intelligence (AI) to aid in the detection of focal liver lesions (FLLs) during ultrasound. An AI system for FLL detection in ultrasound videos was developed. Data in this study were prospectively collected at a university hospital. We applied a two-step training strategy for developing the AI system by using a large collection of ultrasound snapshot images and frames from full-length ultrasound videos. Detection performance of the AI system was evaluated and then compared to detection performance by 25 physicians including 16 non-radiologist physicians and 9 radiologists. Our dataset contained 446 videos (273 videos with 387 FLLs and 173 videos without FLLs) from 334 patients. The videos yielded 172,035 frames with FLLs and 1,427,595 frames without FLLs for training on the AI system. The AI system achieved an overall detection rate of 89.8% (95%CI: 84.5-95.0) which was significantly higher than that achieved by non-radiologist physicians (29.1%, 95%CI: 21.2-37.0, p < 0.001) and radiologists (70.9%, 95%CI: 63.0-78.8, p < 0.001). Median false positive detection rate by the AI system was 0.7% (IQR: 1.3%). AI system operation speed reached 30-34 frames per second, showing real-time feasibility. A further study to demonstrate whether the AI system can assist operators during ultrasound examinations is warranted.
尽管广泛提供了用于肝细胞癌监测的超声机，但在偏远地区进行超声检查的专家放射科医生数量不足仍然是监测的主要障碍。我们证明了人工智能 (AI) 在超声中辅助检测局灶性肝病变 (FLL) 的可行性。开发了一种用于超声视频中 FLL 检测的 AI 系统。本研究的数据是在一所大学医院前瞻性收集的。我们应用两步训练策略，使用大量超声快照图像和全长超声视频的帧来开发 AI 系统。评估了 AI 系统的检测性能，然后将其与 25 名医生（包括 16 名非放射科医生和 9 名放射科医生）的检测性能进行比较。我们的数据集包含 334 名患者的 446 个视频（273 个视频中有 387 个 FLL 和 173 个没有 FLL 的视频）。这些视频产生了 172,035 个有 FLL 的帧和 1,427,595 个没有 FLL 的帧，用于 AI 系统的训练。AI 系统的总体检测率为 89.8%（95%CI：84.5-95.0），明显高于非放射科医生（29.1%，95%CI：21.2-37.0，p<0.001）和放射科医生（70.9%，95%CI：63.0-78.8，p<0.001）的检测率。AI 系统的平均假阳性检出率为 0.7%（IQR：1.3%）。AI 系统的操作速度达到 30-34 帧/秒，具有实时可行性。需要进一步的研究来证明 AI 系统是否可以在超声检查期间协助操作人员。
