Multi-Scale Modeling to Predict Long-Term Growth and Remodeling of Skin in Response to Stretch

多尺度建模预测皮肤对拉伸反应的长期生长和重塑

• 批准号: 10605576
• 负责人: Adrian Buganza Tepole
• 金额: $ 18.93万
• 依托单位: PURDUE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-07-15 至 2024-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10605576
• 关键词: 3-Dimensional; Administrative Supplement; Area; Artificial Intelligence; Award; Biological Process; Biomedical Engineering; Biophysics; Breast Cancer Treatment; Calibration; Cell Proliferation; Cells; Characteristics; Clinical; Collection; Complement; Computer Models; Cosmetics; Coupled; Data; Development; Educational workshop; Engineering; Family suidae; Fostering; Future; Geometry; Growth; Human; Lead; Learning; Machine Learning; Malignant Neoplasms; Mammaplasty; Mastectomy; Measurement; Mechanics; Mentors; Methods; Modeling; Outcome; Paper; Parents; Pathway interactions; Patients; Pattern; Physics; Process; Protocols documentation; Regulatory Element; Research; Role; Science; Scientist; Shapes; Signal Transduction; Skin; Stretching; System; Techniques; Testing; Three-Dimensional Image; Time; Tissue Expanders; Tissue Expansion; Tissue imaging; Tissues; Training; Translating; Universities; Vision; Woman; Work; base; biological systems; biomedical imaging; course sequence; data modeling; data-driven model; design; diversity and equity; experimental study; graduate student; image registration; improved; learning strategy; machine learning framework; machine learning model; mechanical signal; minority student; multi-scale modeling; neural network; parent grant; parent project; porcine model; response; simulation; standard care; success; tool

Project Summary Mastectomy continues to be a standard of treatment for breast cancer, the most common cancer in women. Tissue expansion (TE) is the preferred technique for breast reconstruction after mastectomy. Unfortunately, the rate of complications with TE for breast reconstruction can be 15% or higher, not including poor cosmetic outcomes. In the parent grant we are working with our proposed porcine model of TE, as well as our computational modeling framework of skin growth. As part of the parent award, our work has shown that skin growth rate is proportional to the amount of deformation, and that this process has a characteristic time constant on the order of a few days. We have further looked into the cellular mechanisms that drive skin growth and identified key mechanotranduction pathways that lead to increased cell proliferation. We have also started collection of patient data to translate the findings from the porcine model to human patients. On the other hand, machine learning (ML) has permeated engineering sciences, enabling analysis of biological processes that would otherwise be impossible with traditional approaches. In particular, we have been at the forefront of applying ML tools to our experimental data and computational models of skin growth in TE. In this Supplement proposal we will further develop ML tools to identify the signaling network dynamics that best explain the mechanisms by which cells adapt to mechanical cues (Aim S1); we will create image-registration frameworks using physics-informed neural networks to process 3D images from the clinical setting in which precise measurement of tissue deformation is challenging (Aim S2); and we will establish ML optimization frameworks to design TE protocols that can lead to desired outcomes in terms of time and pattern of skin growth (Aim S3). In parallel to the research objectives, this Supplement will establish a sequence of courses and workshops to mentor senior graduate students and postdoctoral scholars and foster Diversity, Equity, Inclusion and Accessibility (DEIA) at Purdue University.

项目摘要 乳腺癌是女性最常见的癌症，乳房切除术仍然是治疗乳腺癌的标准。 组织扩张术（TE）是乳房切除术后乳房重建的首选技术。不幸的是， TE乳房再造的并发症发生率可能为15%或更高，不包括外观不良 结果。在母基金中，我们正在研究我们提出的猪TE模型，以及我们的 皮肤生长的计算建模框架。作为父母奖的一部分，我们的工作表明，皮肤 生长速率与变形量成正比，并且该过程具有特征时间 持续几天我们已经进一步研究了驱动皮肤的细胞机制， 生长，并确定了导致细胞增殖增加的关键机械传导途径。我们还 开始收集患者数据，以将猪模型的发现转化为人类患者。上 另一方面，机器学习（ML）已经渗透到工程科学中，使生物分析成为可能。 这是传统方法无法实现的。特别是，我们一直在 将ML工具应用于TE中皮肤生长的实验数据和计算模型的最前沿。在这 补充建议，我们将进一步开发ML工具，以确定最佳的信令网络动态 解释细胞适应机械提示的机制（目标S1）;我们将创建图像配准 框架使用物理信息神经网络来处理来自临床环境的3D图像，其中 组织变形的精确测量具有挑战性（目标S2）;我们将建立ML优化 设计TE协议的框架，可以在时间和皮肤模式方面产生预期的结果 增长（目标S3）。在研究目标的同时，本补编将建立一个课程序列 和讲习班，指导高年级研究生和博士后学者，促进多样性，公平， 普渡大学的包容性和可访问性（DEIA）。