Graduate School of Medicine, Kyoto University, Kyoto, Japan.
SIT Research Laboratories, Shibaura Institute of Technology, 3-7-5 Toyosu, Koto-ku, Tokyo, 135-8548, Japan.
J Med Ultrason (2001). 2023 Apr;50(2):167-176. doi: 10.1007/s10396-023-01301-2. Epub 2023 Apr 4.
In recent years, efforts to apply artificial intelligence (AI) to the medical field have been growing. In general, a vast amount of high-quality training data is necessary to make great AI. For tumor detection AI, annotation quality is important. In diagnosis and detection of tumors using ultrasound images, humans use not only the tumor area but also the surrounding information, such as the back echo of the tumor. Therefore, we investigated changes in detection accuracy when changing the size of the region of interest (ROI, ground truth area) relative to liver tumors in the training data for the detection AI.
We defined D/L as the ratio of the maximum diameter (D) of the liver tumor to the ROI size (L). We created training data by changing the D/L value, and performed learning and testing with YOLOv3.
Our results showed that the detection accuracy was highest when the training data were created with a D/L ratio between 0.8 and 1.0. In other words, it was found that the detection accuracy was improved by setting the ground true bounding box for detection AI training to be in contact with the tumor or slightly larger. We also found that when the D/L ratio was distributed in the training data, the wider the distribution, the lower the detection accuracy.
Therefore, we recommend that the detector be trained with the D/L value close to a certain value between 0.8 and 1.0 for liver tumor detection from ultrasound images.
近年来，将人工智能（AI）应用于医学领域的努力一直在增加。一般来说，要制作出优秀的 AI，需要大量的高质量训练数据。对于肿瘤检测 AI，注释质量很重要。在使用超声图像进行肿瘤诊断和检测时，人类不仅使用肿瘤区域，还使用周围信息，例如肿瘤的背回波。因此，我们研究了在训练数据中改变 ROI（感兴趣区域，真实区域）相对于肝肿瘤的大小对肿瘤检测 AI 的检测准确性的变化。
我们将 D/L 定义为肝肿瘤的最大直径（D）与 ROI 大小（L）的比值。我们通过改变 D/L 值来创建训练数据，并使用 YOLOv3 进行学习和测试。
我们的结果表明，当训练数据的 D/L 比值在 0.8 到 1.0 之间时，检测准确性最高。换句话说，通过将检测 AI 训练的真实边界框设置为与肿瘤接触或稍大一些，发现可以提高检测准确性。我们还发现，当 D/L 比值分布在训练数据中时，分布越广，检测准确性越低。
因此，我们建议在训练肝肿瘤超声图像检测的检测器时，将 D/L 值接近 0.8 到 1.0 之间的某个值。
