Browsing by Author "Karam, Jose A."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A generative adversarial approach to facilitate archival-quality histopathologic diagnoses from frozen tissue sections
    (Elsevier, 2022) Falahkheirkhah, Kianoush; Guo, Tao; Hwang, Michael; Tamboli, Pheroze; Wood, Christopher G.; Karam, Jose A.; Sircar, Kanishka; Bhargava, Rohit; Pathology and Laboratory Medicine, School of Medicine
    In clinical diagnostics and research involving histopathology, formalin-fixed paraffin-embedded (FFPE) tissue is almost universally favored for its superb image quality. However, tissue processing time (>24 h) can slow decision-making. In contrast, fresh frozen (FF) processing (<1 h) can yield rapid information but diagnostic accuracy is suboptimal due to lack of clearing, morphologic deformation and more frequent artifacts. Here, we bridge this gap using artificial intelligence. We synthesize FFPE-like images ("virtual FFPE") from FF images using a generative adversarial network (GAN) from 98 paired kidney samples derived from 40 patients. Five board-certified pathologists evaluated the results in a blinded test. Image quality of the virtual FFPE data was assessed to be high and showed a close resemblance to real FFPE images. Clinical assessments of disease on the virtual FFPE images showed a higher inter-observer agreement compared to FF images. The nearly instantaneously generated virtual FFPE images can not only reduce time to information but can facilitate more precise diagnosis from routine FF images without extraneous costs and effort.
  • Thumbnail Image
    Item
    Predictive Nomogram for Recurrence following Surgery for Nonmetastatic Renal Cell Cancer with Tumor Thrombus
    (Elsevier, 2017-10) Abel, E. Jason; Masterson, Timothy A.; Karam, Jose A.; Master, Viraj A.; Margulis, Vitaly; Hutchinson, Ryan; Lorentz, C. Adam; Bloom, Evan; Bauman, Tyler M.; Wood, Christopher G.; Blute, Michael L., Jr.; Department of Urology, School of Medicine
    Purpose Following surgery for nonmetastatic renal cell carcinoma with tumor thrombus the risk of recurrence is significant but variable among patients. The purpose of this study was to develop and validate a predictive nomogram for individual estimation of recurrence risk following surgery for renal cell carcinoma with venous tumor thrombus. Materials and Methods Comprehensive data were collected on patients with nonmetastatic renal cell carcinoma and thrombus treated at a total of 5 institutions from 2000 to 2013. Independent predictors of recurrent renal cell carcinoma from a competing risks analysis were developed into a nomogram. Predictive accuracy was compared between the development and validation cohorts, and between the nomogram and the UISS (UCLA Integrated Staging System, SSIGN (Stage, Size, Grade and Necrosis) and Sorbellini models. Results A total of 636 patients were analyzed, including the development cohort of 465 and the validation cohort of 171. Independent predictors, including tumor diameter, body mass index, preoperative hemoglobin less than the lower limit of normal, thrombus level, perinephric fat invasion and nonclear cell histology, were developed into a nomogram. Estimated 5-year recurrence-free survival was 49% overall. Five-year recurrence-free survival in patients with 0, 1, 2 and more than 2 risk factors was 77%, 53%, 47% and 20%, respectively. Predictive accuracy was similar in the development and validation cohorts (AUC 0.726 and 0.724, respectively). Predictive accuracy of the thrombus nomogram was higher than that of the UISS (AUC 0.726 vs 0.595, p = 0.001), SSIGN (AUC 0.713 vs 0.612, p = 0.04) and Sorbellini models (AUC 0.709 vs 0.638, p = 0.02). Conclusions We present a predictive nomogram for postoperative recurrence in patients with nonmetastatic renal cell carcinoma with venous thrombus. Improving individual postoperative risk assessment may allow for better design and analysis of future adjuvant clinical trials.