Skip to main content

Advertisement

Log in

Image-based clinical decision support for transrectal ultrasound in the diagnosis of prostate cancer: comparison of multiple logistic regression, artificial neural network, and support vector machine

  • Urogenital
  • Published:
European Radiology Aims and scope Submit manuscript

Abstract

Purpose

We developed a multiple logistic regression model, an artificial neural network (ANN), and a support vector machine (SVM) model to predict the outcome of a prostate biopsy, and compared the accuracies of each model.

Method

One thousand and seventy-seven consecutive patients who had undergone transrectal ultrasound (TRUS)-guided prostate biopsy were enrolled in the study. Clinical decision models were constructed from the input data of age, digital rectal examination findings, prostate-specific antigen (PSA), PSA density (PSAD), PSAD in transitional zone, and TRUS findings. The patients were divided into the training and test groups in a randomized fashion. Areas under the receiver operating characteristic (ROC) curve (AUC, Az) were calculated to summarize the overall performance of each decision model for the task of prostate cancer prediction.

Results

The Az values of the ROC curves for the use of multiple logistic regression analysis, ANN, and the SVM were 0.768, 0.778, and 0.847, respectively. Pairwise comparison of the ROC curves determined that the performance of the SVM was superior to that of the ANN or the multiple logistic regression model.

Conclusion

Image-based clinical decision support models allow patients to be informed of the actual probability of having a prostate cancer.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

References

  1. Stephan C, Jung K, Cammann H et al (2002) An artificial neural network considerably improves the diagnostic power of percent free prostate-specific antigen in prostate cancer diagnosis: results of a 5-year investigation. Int J Cancer 99:466–473

    Article  CAS  PubMed  Google Scholar 

  2. Catalona WJ, Richie JP, Ahmann FR et al (1994) Comparison of digital rectal examination and serum prostate specific antigen in the early detection of prostate cancer: results of a multicenter clinical trial of 6,630 men. J Urol 151:1283–1290

    CAS  PubMed  Google Scholar 

  3. Woolf SH (1995) Screening for prostate cancer with prostate-specific antigen. An examination of the evidence. N Engl J Med 333:1401–1405

    Article  CAS  PubMed  Google Scholar 

  4. Catalona WJ, Smith DS, Ratliff TL et al (1991) Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med 324:1156–1161

    Article  CAS  PubMed  Google Scholar 

  5. Arcangeli CG, Ornstein DK, Keetch DW, Andriole GL (1997) Prostate-specific antigen as a screening test for prostate cancer. The United States experience. Urol Clin North Am 24:299–306

    Article  CAS  PubMed  Google Scholar 

  6. Richie JP, Catalona WJ, Ahmann FR et al (1993) Effect of patient age on early detection of prostate cancer with serum prostate-specific antigen and digital rectal examination. Urology 42:365–374

    Article  CAS  PubMed  Google Scholar 

  7. Catalona WJ, Smith DS, Ornstein DK (1997) Prostate cancer detection in men with serum PSA concentrations of 2.6 to 4.0 ng/mL and benign prostate examination. Enhancement of specificity with free PSA measurements. Jama 277:1452–1455

    Article  CAS  PubMed  Google Scholar 

  8. Lawrentschuk N, Fleshner n (2009) the role of magnetic resonance imaging in targeting prostate cancer in patients with previous negative biopsies and elevated prostate-specific antigen levels. BJU Int 103:730–733

    Article  PubMed  Google Scholar 

  9. Suzuki H, Komiya A, Kamiya N et al (2006) Development of a nomogram to predict probability of positive initial prostate biopsy among Japanese patients. Urology 67:131–136

    Article  PubMed  Google Scholar 

  10. Kamoi K, Babaian RJ (1999) Advances in the application of prostate-specific antigen in the detection of early-stage prostate cancer. Semin Oncol 26:140–149

    CAS  PubMed  Google Scholar 

  11. Catalona WJ, Partin AW, Slawin KM et al (1998) Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. Jama 279:1542–1547

    Article  CAS  PubMed  Google Scholar 

  12. Suzuki H, Akakura K, Igarashi T et al (2004) Clinical usefulness of serum antip53 antibodies for prostate cancer detection: a comparative study with prostate specific antigen parameters. J Urol 171:182–186

    Article  CAS  PubMed  Google Scholar 

  13. Snow PB, Smith DS, Catalona WJ (1994) Artificial neural networks in the diagnosis and prognosis of prostate cancer: a pilot study. J Urol 152:1923–1926

    CAS  PubMed  Google Scholar 

  14. Stephan C, Cammann H, Semjonow A et al (2002) Multicenter evaluation of an artificial neural network to increase the prostate cancer detection rate and reduce unnecessary biopsies. Clin Chem 48:1279–1287

    CAS  PubMed  Google Scholar 

  15. Karakiewicz PI, Benayoun S, Kattan MW et al (2005) Development and validation of a nomogram predicting the outcome of prostate biopsy based on patient age, digital rectal examination and serum prostate specific antigen. J Urol 173:1930–1934

    Article  PubMed  Google Scholar 

  16. Chun FK, Briganti A, Graefen M et al (2007) Development and external validation of an extended 10-core biopsy nomogram. Eur Urol 52:436–444

    Article  PubMed  Google Scholar 

  17. Finne P, Finne R, Bangma C et al (2004) Algorithms based on prostate-specific antigen (PSA), free PSA, digital rectal examination and prostate volume reduce false-positive PSA results in prostate cancer screening. Int J Cancer 111:310–315

    Article  CAS  PubMed  Google Scholar 

  18. Walz J, Graefen M, Chun FK et al (2006) High incidence of prostate cancer detected by saturation biopsy after previous negative biopsy series. Eur Urol 50:498–505

    Article  PubMed  Google Scholar 

  19. Nam RK, Toi A, Klotz LH et al (2007) Assessing individual risk for prostate cancer. J Clin Oncol 25:3582–3588

    Article  CAS  PubMed  Google Scholar 

  20. Bianco FJ Jr (2006) Nomograms and medicine. Eur Urol 50:884–886

    Article  PubMed  Google Scholar 

  21. Loch T, Leuschner I, Genberg C et al (1999) Artificial neural network analysis (ANNA) of prostatic transrectal ultrasound. Prostate 39:198–204

    Article  CAS  PubMed  Google Scholar 

  22. Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20:173–297

    Google Scholar 

  23. Carlson GD, Calvanese CB, Partin AW (1998) An algorithm combining age, total prostate-specific antigen (PSA), and percent free PSA to predict prostate cancer: results on 4298 cases. Urology 52:455–461

    Article  CAS  PubMed  Google Scholar 

  24. Eastham JA, May R, Robertson JL, Sartor O, Kattan MW (1999) Development of a nomogram that predicts the probability of a positive prostate biopsy in men with an abnormal digital rectal examination and a prostate-specific antigen between 0 and 4 ng/mL. Urology 54:709–713

    Article  CAS  PubMed  Google Scholar 

  25. Potter SR, Horniger W, Tinzl M, Bartsch G, Partin AW (2001) Age, prostate-specific antigen, and digital rectal examination as determinants of the probability of having prostate cancer. Urology 57:1100–1104

    Article  CAS  PubMed  Google Scholar 

  26. Ohori M, Swindle P (2002) Nomograms and instruments for the initial prostate evaluation: the ability to estimate the likelihood of identifying prostate cancer. Semin Urol Oncol 20:116–122

    Article  PubMed  Google Scholar 

  27. Jiang L, Manry MT (2008) Nonlinear networks for classification. ftp://ftp.simtel.net/pub/simtelnet/msdos/calculte. Accessed 13 Mar 2008

  28. Comak E, Arslan A, Turkoglu I (2007) A decision support system based on support vector machines for diagnosis of the heart valve diseases. Comput Biol Med 37:21–27

    Article  PubMed  Google Scholar 

  29. Chang C-C, Lin C-J (2008) LIBSVM–A library for support vector machines. http://www.csie.ntu.edu.tw/~cjlin/libsvm. Accessed 1 Jul 2008

  30. Kattan MW (2002) Nomograms. Introduction. Semin Urol Oncol 20:79–81

    PubMed  Google Scholar 

  31. Porter CR, Gamito EJ, Crawford ED et al (2005) Model to predict prostate biopsy outcome in large screening population with independent validation in referral setting. Urology 65:937–941

    Article  PubMed  Google Scholar 

  32. Oesterling JE, Cooner WH, Jacobsen SJ, Guess HA, Lieber MM (1993) Influence of patient age on the serum PSA concentration. An important clinical observation. Urol Clin North Am 20:671–680

    CAS  PubMed  Google Scholar 

  33. Carter HB, Pearson JD, Metter EJ et al (1992) Longitudinal evaluation of prostate-specific antigen levels in men with and without prostate disease. Jama 267:2215–2220

    Article  CAS  PubMed  Google Scholar 

  34. Richardson TD, Oesterling JE (1997) Age-specific reference ranges for serum prostate-specific antigen. Urol Clin North Am 24:339–351

    Article  CAS  PubMed  Google Scholar 

  35. Benson MC, Whang IS, Olsson CA, McMahon DJ, Cooner WH (1992) The use of prostate specific antigen density to enhance the predictive value of intermediate levels of serum prostate specific antigen. J Urol 147:817–821

    CAS  PubMed  Google Scholar 

  36. Kattan MW, Eastham JA, Wheeler TM et al (2003) Counseling men with prostate cancer: a nomogram for predicting the presence of small, moderately differentiated, confined tumors. J Urol 170:1792–1797

    Article  PubMed  Google Scholar 

  37. Steyerberg EW, Roobol MJ, Kattan MW, van der Kwast TH, de Koning HJ, Schroder FH (2007) Prediction of indolent prostate cancer: validation and updating of a prognostic nomogram. J Urol 177:107–112 discussion 112

    Article  CAS  PubMed  Google Scholar 

  38. Errejon A, Crawford ED, Dayhoff J et al (2001) Use of artificial neural networks in prostate cancer. Mol Urol 5:153–158

    Article  CAS  PubMed  Google Scholar 

  39. Anagnostou T, Remzi M, Djavan B (2003) Artificial neural networks for decision-making in urologic oncology. Rev Urol 5:15–21

    PubMed  Google Scholar 

  40. Babaian RJ, Miyashita H, Evans RB, Ramirez EI (1992) The distribution of prostate specific antigen in men without clinical or pathological evidence of prostate cancer: relationship to gland volume and age. J Urol 147:837–840

    CAS  PubMed  Google Scholar 

  41. Babaian RJ, Fritsche H, Ayala A et al (2000) Performance of a neural network in detectong prostate cancer in the prostate-specific antigen reflex range of 2.5 to 4.0 ng/mL. Urology 56:1000–1006

    Article  CAS  PubMed  Google Scholar 

  42. Remzi M, Anagnostou T, Ravery V et al (2003) An artificial neural network to predict the outcome of repeat prostate biopsies. Urology 62:456–460

    Article  PubMed  Google Scholar 

  43. Remzi M, Djavan B, Hruby S et al (2001) Use of anewly developed artificial neural network(ANN) to select patients for repeat prostatic biopsies. Eur Urol 39:A106

    Google Scholar 

  44. Han M, Snow PB, Brandt JM, Partin AW (2001) Evaluation of artificial neural networks for the prediction of pathologic stage in prostate carcinoma. Cancer 91:1661–1666

    Article  CAS  PubMed  Google Scholar 

  45. Byvatov E, Fechner U, Sadowski J, Schneider G (2003) Comparison of support vector machine and artificial neural network systems for drug/nondrug classification. J Chem Inf Comput Sci 43:1882–1889

    CAS  PubMed  Google Scholar 

  46. Giiler NF, Kocer S (2005) Use of support vector machines and neural network in diagnosis of neuromuscular disorders. J Med Syst 29:271–284. doi:10.1007/s10916-005-5187-4

    Article  Google Scholar 

  47. Chu A, Ahn h, Halwan B et al (2008) A decision support system to facilitate management of patients with acute gastrointestinal bleeding. Artif Intell Med 42:247–259

    Article  PubMed  Google Scholar 

  48. Mavroforakis ME, Georgiou HV, Dimitropoulos N, Cavouras D, Theodoridis S (2006) Mammographic masses characterization based on localized texture and dataset fractal analysis using linear, neural and support vector machine classifiers. Artif Intell Med 37:145–162

    Article  PubMed  Google Scholar 

  49. Veltri RW, Chaudhari M, Miller MC, Poole EC, O’Dowd GJ, Partin AW (2002) Comparison of logistic regression and neural net modeling for prediction of prostate cancer pathologic stage. Clin Chem 48:1828–1834

    CAS  PubMed  Google Scholar 

  50. Cho J, Kim S, Lee S (2000) Peripheral hypoechoic lesions of the prostate: evaluation with color and power Doppler ultrasound. Eur Urol 37:443–448

    Article  CAS  PubMed  Google Scholar 

  51. Lee HJ, Kim KG, Lee SE et al (2006) Role of transrectal ultrasonography in the prediction of prostate cancer: artificial neural network analysis. J Ultrasound Med 25:815–821 quiz 822–814

    PubMed  Google Scholar 

  52. Rabbani F, Stroumbakis N, Kava B, Cookson M, Fair W (1998) Incidence et clinical significance of false-negative sextant prostate biopsies. J Urol 159:1247–1250

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

This work was supported by the Interdisciplinary Research Initiatives Program by College of Engineering and College of Medicine, Seoul National University (2007, 800–20070086).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Sung Il Hwang.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Lee, H.J., Hwang, S.I., Han, Sm. et al. Image-based clinical decision support for transrectal ultrasound in the diagnosis of prostate cancer: comparison of multiple logistic regression, artificial neural network, and support vector machine. Eur Radiol 20, 1476–1484 (2010). https://doi.org/10.1007/s00330-009-1686-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-009-1686-x

Keywords

Navigation