Valeri Federico, Bartolucci Maurizio, Cantoni Elena, Carpi Roberto, Cisbani Evaristo, Cupparo Ilaria, Doria Sandra, Gori Cesare, Grigioni Mauro, Lasagni Lorenzo, Marconi Alessandro, Mazzoni Lorenzo Nicola, Miele Vittorio, Pradella Silvia, Risaliti Guido, Sanguineti Valentina, Sona Diego, Vannucchi Letizia, Taddeucci Adriana
Università degli Studi di Firenze, Dipartimento di Fisica e Astronomia, Florence, Italy.
Università degli Studi di Firenze, Scuola di Scienze della Salute Umana, Florence, Italy.
J Med Imaging (Bellingham). 2023 Feb;10(Suppl 1):S11904. doi: 10.1117/1.JMI.10.S1.S11904. Epub 2023 Mar 7.
The aim of this work is the development and characterization of a model observer (MO) based on convolutional neural networks (CNNs), trained to mimic human observers in image evaluation in terms of detection and localization of low-contrast objects in CT scans acquired on a reference phantom. The final goal is automatic image quality evaluation and CT protocol optimization to fulfill the ALARA principle.
Preliminary work was carried out to collect localization confidence ratings of human observers for signal presence/absence from a dataset of 30,000 CT images acquired on a PolyMethyl MethAcrylate phantom containing inserts filled with iodinated contrast media at different concentrations. The collected data were used to generate the labels for the training of the artificial neural networks. We developed and compared two CNN architectures based respectively on Unet and MobileNetV2, specifically adapted to achieve the double tasks of classification and localization. The CNN evaluation was performed by computing the area under localization-ROC curve (LAUC) and accuracy metrics on the test dataset.
The mean of absolute percentage error between the LAUC of the human observer and MO was found to be below 5% for the most significative test data subsets. An elevated inter-rater agreement was achieved in terms of S-statistics and other common statistical indices.
Very good agreement was measured between the human observer and MO, as well as between the performance of the two algorithms. Therefore, this work is highly supportive of the feasibility of employing CNN-MO combined with a specifically designed phantom for CT protocol optimization programs.
本研究旨在开发并表征一种基于卷积神经网络（CNN）的模型观察者（MO），该观察者经过训练，可在对参考体模采集的CT扫描图像中低对比度物体的检测和定位方面，模仿人类观察者进行图像评估。最终目标是实现自动图像质量评估和CT协议优化，以符合合理可行尽量低（ALARA）原则。
初步工作是从一个包含30000张CT图像的数据集收集人类观察者对信号存在/不存在的定位置信度评分，这些图像是在一个聚甲基丙烯酸甲酯体模上采集的，该体模含有填充不同浓度碘化造影剂的插入物。收集到的数据用于生成训练人工神经网络的标签。我们分别基于Unet和MobileNetV2开发并比较了两种CNN架构，它们经过专门调整以实现分类和定位的双重任务。通过计算测试数据集上的定位ROC曲线下面积（LAUC）和准确率指标来进行CNN评估。
对于最具代表性的测试数据子集，发现人类观察者和MO的LAUC之间的绝对百分比误差平均值低于5%。在S统计量和其他常见统计指标方面实现了较高的评分者间一致性。
测量发现人类观察者和MO之间以及两种算法的性能之间具有非常好的一致性。因此，本研究高度支持将CNN-MO与专门设计的体模相结合用于CT协议优化程序的可行性。
