Theoretical Framework to Predict Generalized Contrast-to-Noise Ratios of Photoacoustic Images With Applications to Computer Vision
Theoretical Framework to Predict Generalized Contrast-to-Noise Ratios of Photoacoustic Images With Applications to Computer Vision
复制标题
DOI：
10.1109/tuffc.2022.3169082
复制
发表时间：
2022-06-01
影响因子：
3.6
通讯作者：
Bell, Muyinatu A. Lediju
中科院分区：
文献类型：
--
作者：
Gubbi, Mardava R.;Gonzalez, Eduardo A.;Bell, Muyinatu A. Lediju
在外科手术和介入手术过程中，计算机视觉、机器人驱动和光声成像的成功整合需要精确的光声目标检测能力。这种可检测性传统上是用图像质量指标来评估的，比如对比度、对比噪声比和信噪比（SNR）。然而，当使用这些传统指标时，由于无界值和对阈值等图像处理技术的敏感性，预测目标跟踪性能预期是困难的。广义对比噪声比（gCNR）是最近引入的一种替代目标可探测性度量，以前的工作致力于光声图像适用性的经验证明。在本文中，我们提出了理论方法来建模和预测光声图像的gCNR，并提供了相关的理论框架来分析成像系统参数与计算机视觉任务性能之间的关系。我们的理论gCNR预测是通过基于直方图的gCNR测量来验证的，这些测量来自模拟、实验模型、离体和体内数据集。每个数据集的gCNR预测值与实测值之间的平均绝对误差范围为3.2 × 10（-3）至2.3 × 10(-2)，通道信噪比范围为-40至40 dB，激光能量范围为0.07 μ J至68 mJ。研究了gCNR与激光能量、目标和背景图像参数、目标分割和阈值水平之间的关系。研究结果为预测光声gCNR和视觉伺服分割精度提供了良好的基础。预先手术和介入任务的效率（例如，光声引导手术的能量选择）也可以通过提出的框架得到改善。
The successful integration of computer vision, robotic actuation, and photoacoustic imaging to find and follow targets of interest during surgical and interventional procedures requires accurate photoacoustic target detectability. This detectability has traditionally been assessed with image quality metrics, such as contrast, contrast-to-noise ratio, and signal-to-noise ratio (SNR). However, predicting target tracking performance expectations when using these traditional metrics is difficult due to unbounded values and sensitivity to image manipulation techniques like thresholding. The generalized contrast-to-noise ratio (gCNR) is a recently introduced alternative target detectability metric, with previous work dedicated to empirical demonstrations of applicability to photoacoustic images. In this article, we present theoretical approaches to model and predict the gCNR of photoacoustic images with an associated theoretical framework to analyze relationships between imaging system parameters and computer vision task performance. Our theoretical gCNR predictions are validated with histogram-based gCNR measurements from simulated, experimental phantom, ex vivo, and in vivo datasets. The mean absolute errors between predicted and measured gCNR values ranged from 3.2 x 10(-3) to 2.3 x 10(-2) for each dataset, with channel SNRs ranging -40 to 40 dB and laser energies ranging 0.07 mu J to 68 mJ. Relationships among gCNR, laser energy, target and background image parameters, target segmentation, and threshold levels were also investigated. Results provide a promising foundation to enable predictions of photoacoustic gCNR and visual servoing segmentation accuracy. The efficiency of precursory surgical and interventional tasks (e.g., energy selection for photoacoustic-guided surgeries) may also be improved with the proposed framework.