Tensor-based Dictionary Learning for Imaging Biomarkers

 描述(申请人提供)：作为系统生物医学的中心概念，生物标记物是生理和病理状态和活动的多尺度、多样化和相互关联的指示物。在过去的十年里，这一领域的研究一直是活跃和令人兴奋的，包括基于成像生物标志物的成像信息学。在这种情况下，正在进行全基因组关联研究，以建立遗传型和表型生物标记物之间的基本联系，但主要的挑战是，在这一方向上的进展与人们普遍预期的相去甚远。一个关键的观察是，尽管来自基因组测序和表观遗传分析的数据呈爆炸式增长，但在大多数情况下，医学图像特征仍然是主观的或仅以经典方式定义，这似乎是基因型和表型世界之间不合理的失衡。肺癌筛查是一种新兴的CT应用，也是识别成像生物标志物的机会。像其他癌症一样，肺癌不是一种疾病，而是许多疾病。在每个患者中，甚至在每个肿瘤部位，它都是不同的，具有压倒性的非线性和动力学。很明显，要赢得与肺癌的战斗，需要全面、适应性和个性化的治疗方法。为了与这一大局保持一致，对复杂而不是简单的生物标志物的研究不仅有助于而且在癌症研究中也是必要的，成像信息学必须通过丰富的活体成像数据对生物标志物进行专属和智能挖掘，以便建立相关和预测模型。R21项目背后的一般假设是，新的表型信息可以从断层扫描数据中解锁，从而显著提高肺癌CT筛查的敏感性和特异性。该项目的总体目标是开发一种基于张量的字典学习方法来提取CT成像生物标记物，并优化基于张量的局部线性嵌入来使用这些生物标记物来区分CT肺部筛查结果。该项目的主要创新之处在于将张量分解、字典学习、压缩传感、低剂量重建、机器学习、局部线性嵌入、超级计算和大数据挖掘协同集成到一种全新的成像信息学方法中，可以被视为类似于基因组测序的“现象组测序”。在这个项目成功完成后，已识别的成像生物标志物将被证明有助于显著降低肺部CT扫描的假阳性率，同时保持假阴性率不变。它还将有助于准确地对肺癌进行分期，并非侵入性地监测癌症的进展和治疗反应。同样重要的是这个项目的技术意义。如果成立，将对整个影像信息学领域产生持久的影响。

 DESCRIPTION (provided by applicant): As a central concept in systems biomedicine, biomarkers are multi-scale, diverse, and inter-connected indicators of physiological and pathological states and activities. Over the past decade, the research in this area has been active and exciting, including imaging informatics based on imaging biomarkers. In this context, the genome-wide association studies are being performed to establish fundamental links between genotypic and phenotypic biomarkers but a prime challenge is that progress along this direction has been far from what was widely expected. A critical observation is that while data are exploding from genome sequencing and epigenetic analysis, in most cases medical image features are still subjective or only defined in classic fashions, which seems an unreasonable imbalance between genotypic and phenotypic worlds. Lung cancer screening is an emerging CT application and an opportunity to identify imaging biomarkers. Like other cancers, lung cancer is not one but many diseases. It is different in each patient and even in each tumor site with overwhelming nonlinearity and dynamics. It is crystal-clear that comprehensive, adaptive and individualized therapies are needed to win the battle against lung cancer. Being consistent to this big picture, research on sophisticated, instead of simplistic, biomarkers is not only helpful but also necessary in cancer research, and imaging informatics must perform exclusive and intelligent mining through rich in vivo imaging data for biomarkers so that correlative and predictive models could be established. The general hypothesis behind this R21 project is that new phenotypic information can be unlocked in tomographic data to improve sensitivity and specificity significantly in lung cancer CT screening. The overall goal of this project is to develp a tensor-based dictionary learning approach for extraction of CT imaging biomarkers, and optimize a tensor-based locally linear embedding to use these biomarkers for differentiation between CT lung screening results. The major innovation of this project is to synergistically integrate tensor decomposition, dictionary learning, compressive sensing, low-dose reconstruction, machine learning, locally linear embedding, super-computing and big data mining into a brand-new imaging informatics approach, which can be viewed as "phenome sequencing" in analog of genome sequencing. Upon the successful completion of this project, the identified imaging biomarkers will have been demonstrated instrumental in reducing the false positive rate significantly for lung CT scans while the false negative rate is kept constant.It will also help accurately stage lung cancers and non-invasively monitor cancer progression and therapeutic response. Equally important is the technical significance of this project. If it is established, a lasting impact will be generated on the field of imaging informatics at large.