[Computerized supported transrectal ultrasound (C-TRUS) in the diagnosis of prostate cancer]

Urologe A. 2004 Nov;43(11):1377-84. doi: 10.1007/s00120-004-0710-7.
[Article in German]


In the diagnosis of prostate cancer digital rectal examination and transrectal ultrasound (TRUS) are the most utilized methods for clinical evaluation. However, both methods are not able to differentiate between benign and malignant findings with a high amount of certainty. Nevertheless, TRUS is an excellent tool to guide biopsies in practically any region of the prostate. The most significant problem of visual TRUS interpretation is the lack of specificity, especially being an inexperienced user. In order to enhance the diagnostic capabilities of TRUS we developed a computerized analysis of the TRUS signal information (C-TRUS/ANNA), which was validated by the pathohistologic findings of radical prostatectomies. The question was asked: Can C-TRUS detect cancer that has been missed by even multiple systematic biopsies? The entrance criteria was prior negative systematic random biopsies regardless of number of biopsy sessions or number of individual biopsy cores. Five C-TRUS subvisual algorithms were utilized to evaluate the information of the ultrasound signal. The most suspicious regions were marked by C-TRUS and biopsied by guiding a needle into that specific location. In this study 132 with a history of 6-72 negative systematic random biopsies (median: 12 cores) were evaluated by C-TRUS. The PSA ranged from 3.1-36 ng/ml with a median of 9.01 ng/ml. C-TRUS detected in 66 (50%) of these 132 patients cancer by targeted biopsies. In thes 66 men the median number of negative biopsy sessions were two and a median of 12 biopsy cores had been taken. From the literature, we would expect a cancer detection rate in this group with systematic sextant biopsies of about 7%. Only five of the detected carcinomas showed a Gleason Score (GS) of 5, were as 25 had a GS of 6, 22 a GS of 7 and 15 a GS above 7. The results of this prospective clinical trail indicate that C-TRUS is able to identify clinically significant cancers that were missed by even multiple systematic random biopsies. In addition, the concept of searching for strategies that utilize expertise and refinement of imaging modalities is supported rather than just elevating the number of random biopsies (i.e. 141 cores in one session).

Publication types

  • Clinical Trial
  • Comparative Study
  • Randomized Controlled Trial
  • Validation Study

MeSH terms

  • Aged
  • Algorithms*
  • Artificial Intelligence*
  • Equipment Failure Analysis
  • Humans
  • Image Enhancement / methods*
  • Image Interpretation, Computer-Assisted / instrumentation
  • Image Interpretation, Computer-Assisted / methods*
  • Male
  • Middle Aged
  • Pattern Recognition, Automated / methods*
  • Prostatic Neoplasms / diagnostic imaging*
  • Prostatic Neoplasms / pathology
  • Rectum / diagnostic imaging
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Ultrasonography / instrumentation
  • Ultrasonography / methods*