Welcome to the Radiomics Laboratory at USC
The Radiomics Lab is a research group in the Radiology Department at the University of Southern California, Los Angeles. Our multidisciplinary team of radiologists, engineers, researchers, programmers and statisticians are committed to developing image analysis tools, cutting-edge clinical workflows and aiding cost-effective and personalized precision medicine.
With the ever growing capabilities of imaging tumors using various sensor technologies at multiple scales (macro-meso- and micro-scopic) and dimensions (static, 2D, 3D, dynamic etc.), along with technological advances in data extraction and storage, we have been able to extract and analyze a large amount of information relating to tumor behavior and tumor microenvironment than previously possible.
We focus on developing robust computational methods of extracting quantifiable features from medical images encompassing a wide range of imaging modalities and diseases and using it to develop clinical decision support tools to aid diagnosticians to make more informed decisions about the diagnosis as well as prognosis.
Here are some highlights from our more recent work:
We will be at the AUA 2018 (http://www.auameeting2018.com/) .
Please stop by our sessions at the Moscone Center in San Francisco (more details shortly)
- MP63-10 : Development of a clinical decision-support tool for classification of renal masses
- MP72-18 : Association of computed tomography-based radiomic features with epigenetic variation of clear cell renal cell carcinoma using DNA methylation
Distinguishing Fibrosis/Necrosis from Teratoma or Viable Disease in the Retroperitoneum in Post-Chemotherapy, Nonseminomatous Testicular Germ Cell Tumor using Quantitative CT Texture Analysis (Scientific Poster)
Kevin G. King, Sumeet Bhanvadia, Saum Ghodoussipour, Darryl H. Hwang, Bino Varghese, Steven Y. Cen, Siamak Daneshmand, Vinay A. Duddalwar,
Summary: CT texture analysis shows promise in differentiating RP LNs with necrosis/fibrosis from LNs with teratoma or viable malignancy, in post-chemotherapy patients with metastatic testicular NSGT. A larger study is needed for further validation, towards a long-term goal of potentially allowing some patients to avoid PC-RPLND.
CT-based texture characterization of lymphadenopathy in urothelial carcinoma: Prediction of treatment response
Tapas K. Tejura, David Quinn, Bino Varghese, Steven Yong Cen, Darryl Hwang, Frank K. Chen, Bhushan Desai, Ting-Wei Fan, Vinay Duddalwar
Summary: CT-based textural features can differentiate between urothelial cancers which are more likely to respond to treatment.