CT-based Biomarker for Diagnosis of COPD Phenotypes and Disease Progression

描述（由申请人提供）：慢性阻塞性肺疾病（COPD）是一种高度且日益流行的疾病，其特征是不完全可逆的气流限制。它目前是美国第三大死亡原因，有超过1300万人患有这种疾病。肺功能检查用于确定整体肺功能损害；然而，确定肺功能下降的具体原因是必要的，以便选择最佳的治疗策略。我们最近开发了参数反应图（PRM）作为一种基于ct的成像生物标志物，能够诊断COPD的两种主要表型：功能性小气道疾病（fSAD）和肺气肿。这种诊断性生物标志物是放射学和肺部医学实践的游戏规则改变者，但为了充分实现其影响，必须开发一种商业化的、fda批准的诊断分析和报告软件应用程序。我们建议将PRM作为评估和诊断COPD表型的ct成像生物标志物，并将与疾病位置相关的详细空间信息可视化。该提案的目标将集中于开发商业PRM诊断软件应用程序，包括使用从nih赞助的临床试验COPDGene获得的图像和患者信息，验证其作为患者健康状况的替代标记。首先，将PRM算法开发成商业上可行的软件包。在验证商业PRM软件的每个单独组件的输出与密歇根大学PRM发明者的内部分析之后，新软件对194名患者的完整PRM分析将与先前发表的结果进行比较，以确认商业软件提供准确的PRM结果。然后对500例患者的CT图像进行分析，优化PRM分析的输入参数，确保PRM结果准确预测COPD进展。总的来说，预计这项工作的成功结果将显著提高慢性阻塞性肺病的CT成像诊断能力，从而改善患者护理并降低医疗保健系统的成本。

DESCRIPTION (provided by applicant): Chronic obstructive pulmonary disease (COPD) is a highly and increasingly prevalent disorder characterized by incomplete reversible airflow limitations. It is presently the third leading cause of death in the US, with over 13 million peopl suffering from the disease. Pulmonary function tests are used to identify global lung function impairment; however, determining the specific cause of decreased lung function is necessary in order to choose the best treatment strategy. We have recently developed the Parametric Response Map (PRM) as a CT-based imaging biomarker capable of diagnosing the two major phenotypes in COPD: functional small airways disease (fSAD) and emphysema. This diagnostic biomarker is a game-changer for radiological and pulmonary medicine practice, but in order for its impact to be fully realized, a commercial, FDA-approved diagnostic analysis and reporting software application must be developed. We propose to develop PRM as a CT-imaging biomarker for assessment and diagnosis of COPD phenotypes and for visualizing detailed spatial information related to disease location. The goals of this proposal will focus on development of a commercial PRM diagnostic software application, including its validation as a surrogate marker of patient health status, using images and patient information obtained from the NIH-sponsored clinical trial COPDGene. First, the PRM algorithm will be developed into a commercially viable software package. After validating the output of each individual component of the commercial PRM software against the in-house analysis of the inventors of PRM at the University of Michigan, the complete PRM analysis of 194 patients by the new software will be compared to previously published results to confirm that the commercial software gives accurate PRM results. Then the CT images from 500 patients will be analyzed to optimize the input parameters of the PRM analysis to ensure that the PRM results accurately predict COPD progression. Overall, it is anticipated that the successful outcome of this effort will significanty improve the diagnostic capability of CT imaging for COPD, leading to improved patient care and a reduction in cost to the healthcare system.