The James Buchanan Brady Urological Institute
 
 
 
  Biomarker Research Laboratory          
Home
Image Analysis Technology
Proteomics Technology
Algorithms
Biorepository & Informatics
Staff
Publications
Contact Us


ALGORITHMS AND NOMOGRAMS FOR Prostate Cancer

Our work continues on the development and use of computational models (algorithms) that combine clinical (serum/urine biomarkers) and pathologic information obtained from either biopsy or the radical prostatectomy specimens. The term algorithm (pronounced AL-go-rith-um) is a procedure or formula for combining information for solving a complex problem; in our case a prediction of a disease outcome for PCa. In mathematics and computer science, an algorithm usually means a well designed procedure that solves a recurrent decision making problem. The concept is to make predictions of stage as well as risk for recurrence and progression of PCa.  Such programs are being developed and have been applied in research being conducted by the Fisher Biomarker Laboratory.

Danil Makarov - Partin Table 2007 STAGING COMPUTER PROGRAM

partin tables

The Partin Table computer program above utilizes only the clinical stage, PSA and Gleason score of the biopsy as the input to calculate the pathology stage.  

UROSCORE STAGING COMPUTER PROGRAM

Using Logistic Regression and
Program by M. Craig Miller

Patient I.D.
Patient Age Total PSA (ng/ml)
Total # of Cores # Positive Cores
Highest Gleason Score
2 3 4 5 6 7 8 9 10
Total % Tumor Involvement    
Presence of Gleason
Grades 4 and/or 5?
Yes No
Involvement of Mid Core
width >=5% Tumor?
Yes No
Involvement of Base and/or
Mid Core width >=5% Tumor?
Yes No

The UroScore staging computer program that uses age, PSA level, quantitative pathology of the biopsy (i.e. number of positive cores and total % tumor involvement), presence of GG 4 or 5 and involvement of mid core > 5% to predict organ confined disease, capsular penetration, and advanced disease.  Both of these algorithms are based upon the application of non-parametric Logistic Regression statistical methodology.

 

THE FUTURE: A COMPUTATIONAL MULTIVARIATE BIOMARKER APPROACH TO MAKE DISEASE PREDICTIONS.

 



U.S. News and World Report Rankings Badge

 

© The Johns Hopkins University, The Johns Hopkins Hospital, and Johns Hopkins Health System. All rights reserved.| Disclaimer
Email: webmaster@urology.jhu.edu | 600 North Wolfe Street, Baltimore, Maryland 21287