Radiomics is the process of extracting and analyzing mineable, quantitative data from radiographic images. The overarching hypothesis of Radiomics is that image features describing size, shape, and texture reflect underlying tumor pathophysiology and hence, can be developed and qualified as biomarkers for prediction, prognostication or response monitoring. Radiomics is designed to use standard-of-care images, allowing the development and curation of large data sets that are needed for statistical power. In the first cycle of this award, we addressed challenges to all steps in the radiomic pipeline, viz (1) defining the impact of acquisition and image reconstruction on the quality of radiomics data; (2) curating to maintain high data quality; (3) qualifying semi-automated segmentation tools; (4) statistical qualification of radiomic features; (5) developing database sharing tools to allow rapi hypothesis testing; and (6) developing and applying informatics approaches to these datasets. With this pipeline, we identified and validated specific features from CT images that accurately predicted survival in lung cancer patients treated with surgery or chemo-radiation. In this competing continuation, we intend to build on this prior work to incorporate radiomics into a decision support system for post-surgery management of non-small cell lung cancer (NSCLC) patients. NSCLC is the leading cause of cancer deaths worldwide and hence, even incremental improvements in decision support can have a profound impact on patients' lives. We will use and extend the radiomics framework that we have developed to address a compelling and focused question in lung cancer care: whether to treat post-surgery patients with adjuvant chemotherapy (AT). Virtually all NSCLC surgical candidates receive high-quality diagnostic CTs. Early stage NSCLC patients are commonly resected with lobectomy and mediastinal lymph node removal. Of these, up to 35 percent will experience distant recurrence within 5 years. Recurrence can be reduced with AT, yet the decision whether or not to treat is not trivial, as AT is associated with significant morbidities and even mortality. This decision is currently ill-informed by cancer stage alone. There are no predictive models that can accurately identify which patients have the highest likelihood of recurrence, thus requiring most aggressive adjuvant follow-up. Our proposal will address this important clinical problem by development of a Risk-of-Recurrence score with a combination of radiomic, clinical and genomic data curated from over 3,600 patients from two institutions. These data and analytical tools will be shared via unique tools developed for this purpose within the framework of the Quantitative Imaging Network (QIN).

Public Health Relevance

Non-small cell lung cancer (NSCLC) is the leading cause of cancer deaths worldwide and hence, even incremental improvements in the care and treatment of these patients can have a profound impact. Virtually all NSCLC patients receive high-quality diagnostic CT imaging studies to assess extent of disease. The work in the current proposal will analyze these images more deeply than the current standard, and use this 'radiomic' information to help oncologists and patients make a better informed decision whether or not to treat post- surgery with adjuvant chemotherapy (AT). This decision is of profound significance, because, although AT can prevent recurrence, it can also cause significant morbidities and even death.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Research Project--Cooperative Agreements (U01)
Project #
Application #
Study Section
Special Emphasis Panel (ZCA1-TCRB-W (J1))
Program Officer
Tata, Darayash B
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
H. Lee Moffitt Cancer Center & Research Institute
United States
Zip Code
Liu, Shuang; Xie, Yiting; Jirapatnakul, Artit et al. (2017) Pulmonary nodule classification in lung cancer screening with three-dimensional convolutional neural networks. J Med Imaging (Bellingham) 4:041308
Li, Qian; Balagurunathan, Yoganand; Liu, Ying et al. (2017) Comparison Between Radiological Semantic Features and Lung-RADS in Predicting Malignancy of Screen-Detected Lung Nodules in the National Lung Screening Trial. Clin Lung Cancer :
Li, Qian; Kim, Jongphil; Balagurunathan, Yoganand et al. (2017) Imaging features from pretreatment CT scans are associated with clinical outcomes in nonsmall-cell lung cancer patients treated with stereotactic body radiotherapy. Med Phys 44:4341-4349
Shafiq-Ul-Hassan, Muhammad; Zhang, Geoffrey G; Latifi, Kujtim et al. (2017) Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels. Med Phys 44:1050-1062
Tunali, Ilke; Stringfield, Olya; Guvenis, Albert et al. (2017) Radial gradient and radial deviation radiomic features from pre-surgical CT scans are associated with survival among lung adenocarcinoma patients. Oncotarget 8:96013-96026
Yip, Stephen S F; Liu, Ying; Parmar, Chintan et al. (2017) Associations between radiologist-defined semantic and automatically computed radiomic features in non-small cell lung cancer. Sci Rep 7:3519
O'Connor, James P B; Aboagye, Eric O; Adams, Judith E et al. (2017) Imaging biomarker roadmap for cancer studies. Nat Rev Clin Oncol 14:169-186
Li, Qian; Kim, Jongphil; Balagurunathan, Yoganand et al. (2017) CT imaging features associated with recurrence in non-small cell lung cancer patients after stereotactic body radiotherapy. Radiat Oncol 12:158
Liu, Ying; Balagurunathan, Yoganand; Atwater, Thomas et al. (2017) Radiological Image Traits Predictive of Cancer Status in Pulmonary Nodules. Clin Cancer Res 23:1442-1449
Beichel, Reinhard R; Smith, Brian J; Bauer, Christian et al. (2017) Multi-site quality and variability analysis of 3D FDG PET segmentations based on phantom and clinical image data. Med Phys 44:479-496

Showing the most recent 10 out of 74 publications