A non-linear mathematical model using optical sensor to predict heart decellularization efficacy
A non-linear mathematical model using optical sensor to predict heart decellularization efficacy
复制标题
DOI：
10.1038/s41598-019-48659-3
复制
发表时间：
2019-08-21
影响因子：
4.6
通讯作者：
Nogueira, Breno Valentim
中科院分区：
文献类型：
--
作者：
Arruda Pereira, Rayssa Helena;Prado, Adilson Ribeiro;Nogueira, Breno Valentim
脱细胞技术的主要问题之一是其在个体器官中的功效的最终评价的主观性。这个问题可能导致细胞再增殖的再现性受限以及对移植组织的更差的反应。我们的建议是分析心脏在灌注脱细胞过程中产生的光学轮廓，作为评估每个器官脱细胞过程的另一种方法。开发了一种由结构化 LED 光源和可调节底座上的光电探测器组成的装置，用于捕获小鼠心脏灌注过程中的透射光与残留 DNA 含量之间的关系。电压-时间图形记录用于识别非线性数学模型，以区分剩余 DNA 高于（不完全脱细胞）和低于（完全脱细胞）标准化限度的脱细胞。结果表明，无论器官的表观透明度如何，对该过程的时间光学评估都可以在初始阶段预测低效的细胞去除。我们的开放系统还创造了添加不同光电探测器的新可能性，例如用于特定波长、图像采集和支架的物理化学评估，以便从数十项研究中收集不同类型的信息。这些数据经过编译并提交给机器学习技术后，有可能引发组织生物工程研究的指数级进步。
One of the main problems of the decellularization technique is the subjectivity of the final evaluation of its efficacy in individual organs. This problem can result in restricted cell repopulation reproducibility and worse responses to transplant tissues. Our proposal is to analyze the optical profiles produced by hearts during perfusion decellularization, as an additional method for evaluating the decellularization process of each individual organ. An apparatus comprised of a structured LED source and photo detector on an adjustable base was developed to capture the relationship between transmitted light during the perfusion of murine hearts, and residual DNA content. Voltage-time graphic records were used to identify a nonlinear mathematical model to discriminate between decellularizations with remaining DNA above (Incomplete Decellularization) and below (Complete Decellularization) the standardized limits. The results indicate that temporal optical evaluation of the process enables inefficient cell removal to be predicted in the initial stages, regardless of the apparent transparency of the organ. Our open system also creates new possibilities to add distinct photo detectors, such as for specific wavelengths, image acquisition, and physical-chemical evaluation of the scaffold, in order to collect different kinds of information, from dozens of studies. These data, when compiled and submitted to machine learning techniques, have the potential to initiate an exponential advance in tissue bioengineering research.