SAIDI, OLIVIER,VERBEL, DAVID A.,TEVEROVSKIY, MIKHAIL
申请号：
CA2624970
公开号：
CA2624970C
申请日:
2006.10.13
申请国别(地区):
CA
年份:
2015
代理人:
摘要：
Methods and systems are provided that use clinical information, molecularinformation and computer-generated morphometric information in a predictivemodel for predicting the occurrence (e.g., recurrence) of a medical condition,for example, cancer. In an embodiment, a model that predicts prostate cancerrecurrence is provided, where the model is based on features including seminalvesicle involvement, surgical margin involvement, lymph node status, androgenreceptor (AR) staining index of tumor, a morphometric measurement ofepithelial nuclei, and at least one morphometric measurement of stroma. Inanother embodiment, a model that predicts clinical failure post prostatectomyis provided, wherein the model is based on features including biopsy Gleasonscore, lymph node involvement, prostatectomy Gleason score, a morphometricmeasurement of epithelial cytoplasm, a morphometric measurement of epithelialnuclei, a morphometric measurement of stroma, and intensity of androgenreceptor (AR) in racemase (AMACR)-positive epithelial cells.
