Department of Radiology, St Vincent's University Hospital, Dublin, Ireland.
Insight Centre for Data Analytics, UCD, Dublin, Ireland.
Eur Radiol. 2023 Nov;33(11):8376-8386. doi: 10.1007/s00330-023-09769-9. Epub 2023 Jun 7.
Siamese neural networks (SNN) were used to classify the presence of radiopaque beads as part of a colonic transit time study (CTS). The SNN output was then used as a feature in a time series model to predict progression through a CTS.
This retrospective study included all patients undergoing a CTS in a single institution from 2010 to 2020. Data were partitioned in an 80/20 Train/Test split. Deep learning models based on a SNN architecture were trained and tested to classify images according to the presence, absence, and number of radiopaque beads and to output the Euclidean distance between the feature representations of the input images. Time series models were used to predict the total duration of the study.
In total, 568 images of 229 patients (143, 62% female, mean age 57) patients were included. For the classification of the presence of beads, the best performing model (Siamese DenseNET trained with a contrastive loss with unfrozen weights) achieved an accuracy, precision, and recall of 0.988, 0.986, and 1. A Gaussian process regressor (GPR) trained on the outputs of the SNN outperformed both GPR using only the number of beads and basic statistical exponential curve fitting with MAE of 0.9 days compared to 2.3 and 6.3 days (p < 0.05) respectively.
SNNs perform well at the identification of radiopaque beads in CTS. For time series prediction our methods were superior at identifying progression through the time series compared to statistical models, enabling more accurate personalised predictions.
Our radiologic time series model has potential clinical application in use cases where change assessment is critical (e.g. nodule surveillance, cancer treatment response, and screening programmes) by quantifying change and using it to make more personalised predictions.
• Time series methods have improved but application to radiology lags behind computer vision. Colonic transit studies are a simple radiologic time series measuring function through serial radiographs. • We successfully employed a Siamese neural network (SNN) to compare between radiographs at different points in time and then used the output of SNN as a feature in a Gaussian process regression model to predict progression through the time series. • This novel use of features derived from a neural network on medical imaging data to predict progression has potential clinical application in more complex use cases where change assessment is critical such as in oncologic imaging, monitoring for treatment response, and screening programmes.
使用暹罗神经网络（SNN）对不透射线珠的存在进行分类，作为结肠通过时间研究（CTS）的一部分。然后，将 SNN 输出用作时间序列模型中的特征，以预测 CTS 中的进展。
本回顾性研究纳入了 2010 年至 2020 年在一家机构进行 CTS 的所有患者。数据以 80/20 的 Train/Test 比例进行分割。基于 SNN 架构的深度学习模型进行训练和测试，以根据不透射线珠的存在、不存在和数量对图像进行分类，并输出输入图像特征表示之间的欧几里得距离。时间序列模型用于预测研究的总持续时间。
共纳入 229 名患者（143 名，女性占 62%，平均年龄 57 岁）的 568 张图像。对于珠状物存在的分类，表现最佳的模型（使用对比损失训练的暹罗密集网络，冻结权重）达到了 0.988 的准确性、精度和召回率，0.986 和 1。与仅使用珠数的高斯过程回归（GPR）和基本统计指数曲线拟合相比，使用 SNN 输出训练的高斯过程回归（GPR）在 MAE 方面表现更好，分别为 0.9 天和 2.3 天和 6.3 天（p < 0.05）。
SNN 在 CTS 中识别不透射线珠方面表现良好。对于时间序列预测，我们的方法在识别时间序列中的进展方面优于统计模型，从而能够进行更准确的个性化预测。
我们的放射时间序列模型在变化评估至关重要的情况下（例如，结节监测、癌症治疗反应和筛查计划）具有潜在的临床应用，通过量化变化并使用它进行更个性化的预测。
时间序列方法已经得到了改进，但在放射学中的应用仍落后于计算机视觉。结肠通过研究是一种简单的放射时间序列，通过连续的射线照片测量功能。
我们成功地使用暹罗神经网络（SNN）来比较不同时间点的射线照片，然后将 SNN 的输出用作高斯过程回归模型中的特征，以预测时间序列中的进展。
这种在医学成像数据中使用源自神经网络的特征来预测进展的新方法在更复杂的用例中具有潜在的临床应用，在这些用例中，变化评估至关重要，例如肿瘤学成像、治疗反应监测和筛查计划。
