Abstract
Artificial neural networks (ANNs) are mathematical models that are based on biological neural networks and are composed of interconnected groups of artificial neurons. ANNs are used to map and predict outcomes in complex relationships between given 'inputs' and sought-after 'outputs' and can also be used find patterns in datasets. In medicine, ANN applications have been used in cancer diagnosis, staging and recurrence prediction since the mid-1990s, when an enormous effort was initiated, especially in prostate cancer detection. Modern ANNs can incorporate new biomarkers and imaging data to improve their predictive power and can offer a number of advantages as clinical decision making tools, such as easy handling of distribution-free input parameters. Most importantly, ANNs consider nonlinear relationships among input data that cannot always be recognized by conventional analyses. In the future, complex medical diagnostic and treatment decisions will be increasingly based on ANNs and other multivariate models.
Key Points
-
The use of multivariate mathematical models, including artificial neural networks (ANNs), in clinical medicine has increased markedly in the past decade
-
The majority of prostate cancer ANNs are used for diagnosis and staging, providing useful clinical decision support mechanisms to physicians
-
These models have greater specificity than serum PSA and percent free PSA as sole predictors in the diagnosis of prostate cancer
-
The number of ANNs that can be used for prognosis, recurrence risk and prediction of metastasis is limited
-
For the staging of prostate cancer, Bayesian neural networks might be superior to other ANNs and bivariate logistic regression models
This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Sargent, D. J. Comparison of artificial neural networks with other statistical approaches. Cancer 91, 1636–1642 (2001).
Schröder, F. & Kattan, M. W. The comparability of models for predicting the risk of a positive prostate biopsy with prostate-specific antigen alone: a systematic review. Eur. Urol. 54, 274–290 (2008).
Krogh, A. What are artificial neural networks? Nat. Biotechnol. 26, 195–197 (2008).
Finne, P. et al. Predicting the outcome of prostate biopsy in screen-positive men by a multilayer perceptron network. Urology 56, 418–422 (2000).
Horninger, W., Bartsch, G., Snow, P. B., Brandt, J. M. & Partin, A. W. The problem of cutoff levels in a screened population. Cancer 91 (Suppl. 8), 1667–1672 (2001).
Remzi, M. et al. An artificial neural network to predict the outcome of repeat prostate biopsies. Urology 62, 456–460 (2003).
Stephan, C. et al. Multicenter evaluation of an artificial neural network to increase prostate cancer detection rate and reduce unnecessary biopsies. Clin. Chem. 48, 1279–1287 (2002).
Capitanio, U. et al. Predictive models before and after radical prostatectomy. Prostate 70, 1371–1378 (2010).
Shariat, S. F., Karakiewicz, P. I., Suardi, N. & Kattan, M. W. Comparison of nomograms with other methods for predicting outcomes in prostate cancer: a critical analysis of the literature. Clin. Cancer Res. 14, 4400–4407 (2008).
McCulloch, W. S. & Pitts, W. The statistical organization of nervous activity. Biometrics 4, 91–99 (1948).
Hebb, D. O. The Organization of Behavior (John Wiley & Sons, New York, 1949).
Rosenblatt, F. The perceptron: a probabilistic model for information storage and organization in the brain. Psychol. Rev. 65, 386–408 (1958).
Picard, R. R. & Berk, K. N. Data splitting. Am. Stat. 44, 140–147 (1990).
Stephan, C. et al. 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 (2002).
Efron, B. & Tibshirani, R. J. An Introduction to the Bootstrap (Chapman & Hall, New York, 1993).
Tafeit, E. & Reibnegger, G. Artificial neural networks in laboratory medicine and medical outcome prediction. Clin. Chem. Lab. Med. 37, 845–853 (1999).
Dreiseitl, S. & Ohno-Machado, L. Logistic regression and artificial neural network classification models: a methodology review. J. Biomed. Inform. 35, 352–359 (2002).
Anagnostou, T., Remzi, M., Lykourinas, M. & Djavan, B. Artificial neural networks for decision-making in urologic oncology. Eur. Urol. 43, 596–603 (2003).
Rodvold, D. M., McLeod, D. G., Brandt, J. M., Snow, P. B. & Murphy, G. P. Introduction to artificial neural networks for physicians: taking the lid off the black box. Prostate 46, 39–44 (2001).
Cammann, H., Jung, K., Meyer, H. A. & Stephan, C. Avoiding pitfalls in applying prediction models, as illustrated by the example of prostate cancer diagnosis. Clin. Chem. 57, 1490–1498 (2011).
Schwarzer, G. & Schumacher, M. Artificial neural networks for diagnosis and prognosis in prostate cancer. Semin. Urol. Oncol. 20, 89–95 (2002).
Snow, P. B., Smith, D. S. & Catalona, W. J. Artificial neural networks in the diagnosis and prognosis of prostate cancer: a pilot study. J. Urol. 152, 1923–1926 (1994).
Babaian, R. J. et al. Performance of a neural network in detecting prostate cancer in the prostate-specific antigen reflex range of 2.5 to 4.0 ng/mL. Urology 56, 1000–1006 (2000).
Chun, F. K. et al. Initial biopsy outcome prediction—head-to-head comparison of a logistic regression-based nomogram versus artificial neural network. Eur. Urol. 51, 1236–1240 (2007).
Meijer, R. P. et al. The value of an artificial neural network in the decision-making for prostate biopsies. World J. Urol. 27, 593–598 (2009).
Matsui, Y. et al. The use of artificial neural network analysis to improve the predictive accuracy of prostate biopsy in the Japanese population. Jpn J. Clin. Oncol. 34, 602–607 (2004).
Stephan, C. et al. Artificial neural network (ANN) velocity better identifies benign prostatic hyperplasia but not prostate cancer compared with PSA velocity. BMC Urol. 8, 10 (2008).
Stephan, C. et al. Interchangeability of measurements of total and free prostate-specific antigen in serum with 5 frequently used assay combinations: an update. Clin. Chem. 52, 59–64 (2006).
Stephan, C. et al. An artificial neural network for five different assay systems of prostate-specific antigen in prostate cancer diagnostics. BJU Int. 102, 799–805 (2008).
Stephan, C. & Cammann, H. ProstataClass [online], (2008).
Stephan, C. et al. Internal validation of an artificial neural network for prostate biopsy outcome. Int. J. Urol. 17, 62–68 (2010).
Ecke, T. H. et al. Outcome prediction for prostate cancer detection rate with artificial neural network (ANN) in daily routine. Urol. Oncol. 30, 139–144 (2012).
Lawrentschuk, N. et al. Predicting prostate biopsy outcome: artificial neural networks and polychotomous regression are equivalent models. Int. Urol. Nephrol. 43, 23–30 (2011).
Lilja, H., Ulmert, D. & Vickers, A. J. Prostate-specific antigen and prostate cancer: prediction, detection and monitoring. Nat. Rev. Cancer 8, 268–278 (2008).
Catalona, W. J. et al. 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 (1998).
Roddam, A. W. et al. Use of prostate-specific antigen (PSA) isoforms for the detection of prostate cancer in men with a PSA level of 2–10 ng/ml: systematic review and meta-analysis. Eur. Urol. 48, 386–399 (2005).
Stephan, C., Lein, M., Jung, K., Schnorr, D. & Loening, S. A. The influence of prostate volume on the ratio of free to total prostate specific antigen in serum of patients with prostate carcinoma and benign prostate hyperplasia. Cancer 79, 104–109 (1997).
Lilja, H., Oldbring, J., Rannevik, G. & Laurell, C. B. Seminal vesicle-secreted proteins and their reactions during gelation and liquefaction of human semen. J. Clin. Invest. 80, 281–285 (1987).
Stephan, C. et al. Clinical utility of human glandular kallikrein 2 within a neural network for prostate cancer detection. BJU Int. 96, 521–527 (2005).
Stephan, C. et al. Improved prostate cancer detection with a human kallikrein 11 and percentage free PSA-based artificial neural network. Biol. Chem. 387, 801–805 (2006).
Stephan, C. et al. Three new serum markers for prostate cancer detection within a percent free PSA-based artificial neural network. Prostate 66, 651–659 (2006).
Stephan, C. et al. A (−5, −7) proPSA based artificial neural network to detect prostate cancer. Eur. Urol. 50, 1014–1020 (2006).
Catalona, W. J. et al. Serum pro prostate specific antigen improves cancer detection compared to free and complexed prostate specific antigen in men with prostate specific antigen 2 to 4 ng/ml. J. Urol. 170, 2181–2185 (2003).
Sokoll, L. J. et al. [−2]Proenzyme prostate specific antigen for prostate cancer detection: a national cancer institute early detection research network validation study. J. Urol. 180, 539–543 (2008).
Stephan, C. et al. A [−2]proPSA-based artificial neural network significantly improves differentiation between prostate cancer and benign prostatic diseases. Prostate 69, 198–207 (2009).
Guazzoni, G. et al. Prostate-specific antigen (PSA) isoform p2PSA significantly improves the prediction of prostate cancer at initial extended prostate biopsies in patients with total PSA between 2.0 and 10 ng/ml: results of a prospective study in a clinical setting. Eur. Urol. 60, 214–222 (2011).
Jansen, F. H. et al. Prostate-specific antigen (PSA) isoform p2PSA in combination with total PSA and free PSA improves diagnostic accuracy in prostate cancer detection. Eur. Urol. 57, 921–927 (2010).
Lughezzani, G. et al. Development and internal validation of a prostate health index (phi) based nomogram for predicting prostate cancer at extended biopsy. J. Urol. 188, 1144–1150 (2012).
Stephan, C. et al. Multicenter evaluation of [-2]proprostate-specific antigen and the prostate health index for detecting prostate cancer. Clin. Chem. 59, 306–314 (2013).
Stephan, C. et al. Benign prostatic hyperplasia-associated free prostate-specific antigen improves detection of prostate cancer in an artificial neural network. Urology 74, 873–877 (2009).
Hessels, D. et al. DD3(PCA3)-based molecular urine analysis for the diagnosis of prostate cancer. Eur. Urol. 44, 8–15 (2003).
Haese, A. et al. Clinical utility of the PCA3 urine assay in European men scheduled for repeat biopsy. Eur. Urol. 54, 1081–1088 (2008).
Chun, F. K. et al. Prostate cancer gene 3 (PCA3): development and internal validation of a novel biopsy nomogram. Eur. Urol. 56, 659–667 (2009).
Ankerst, D. P. et al. Predicting prostate cancer risk through incorporation of prostate cancer gene 3. J. Urol. 180, 1303–1308 (2008).
Ferro, M. et al. Predicting prostate biopsy outcome: prostate health index (phi) and prostate cancer antigen 3 (PCA3) are useful biomarkers. Clin. Chim. Acta 413, 1274–1278 (2012).
Stephan, C. et al. Comparative assessment of urinary prostate cancer antigen 3 and TMPRSS2:ERG gene fusion with the serum [−2]proprostate-specific antigen-based Prostate Health Index for detection of prostate cancer. Clin. Chem. 59, 280–288 (2013).
Aubin, S. M. et al. Improved prediction of prostate biopsy outcome using PCA3, TMPRSS2:ERG gene fusion and serum PSA [abstract 2105]. J. Urol. 179 (Suppl.), 725 (2008).
Groskopf, J. et al. Feasibility and clinical utility of a TMPRSS2:ERG gene fusion urine test [abstract 3]. Eur. Urol. 8 (Suppl.), 195 (2009).
Tomlins, S. A. et al. Urine TMPRSS2:ERG fusion transcript stratifies prostate cancer risk in men with elevated serum PSA. Sci. Transl. Med. 3, 94ra72 (2011).
Panebianco, V. et al. Conventional imaging and multiparametric magnetic resonance (MRI, MRS, DWI, MRP) in the diagnosis of prostate cancer. Q. J. Nucl. Med. Mol. Imaging 56, 331–342 (2012).
Lee, F. et al. Transrectal ultrasound in the diagnosis of prostate cancer: location, echogenicity, histopathology, and staging. Prostate 7, 117–129 (1985).
Shinohara, K., Scardino, P. T., Carter, S. S. & Wheeler, T. M. Pathologic basis of the sonographic appearance of the normal and malignant prostate. Urol. Clin. North Am. 16, 675–691 (1989).
Ronco, A. L. & Fernández, R. Improving ultrasonographic diagnosis of prostate cancer with neural networks. Ultrasound Med. Biol. 25, 729–733 (1999).
Loch, T. et al. Artificial neural network analysis (ANNA) of prostatic transrectal ultrasound. Prostate 39, 198–204 (1999).
Lee, H. J. et al. Role of transrectal ultrasonography in the prediction of prostate cancer: artificial neural network analysis. J. Ultrasound Med. 25, 815–821 (2006).
Poulakis, V. et al. Preoperative neural network using combined magnetic resonance imaging variables, prostate-specific antigen, and Gleason score for predicting prostate cancer biochemical recurrence after radical prostatectomy. Urology 64, 1165–1170 (2004).
Poulakis, V. et al. Preoperative neural network using combined magnetic resonance imaging variables, prostate specific antigen and Gleason score to predict prostate cancer stage. J. Urol. 172, 1306–1310 (2004).
Poulakis, V. et al. Preoperative neural network using combined magnetic resonance imaging variables, prostate specific antigen, and Gleason score to predict prostate cancer recurrence after radical prostatectomy. Eur. Urol. 46, 571–578 (2004).
Poulakis, V. et al. Preoperative neural network using combined magnetic resonance imaging variables, prostate-specific antigen, and Gleason score to predict positive surgical margins. Urology 64, 516–521 (2004).
Porter, C. R. & Crawford, E. D. Combining artificial neural networks and transrectal ultrasound in the diagnosis of prostate cancer. Oncology (Williston. Park) 17, 1395–1399 (2003).
Naguib, R. N., Robinson, M. C., Neal, D. E. & Hamdy, F. C. Neural network analysis of combined conventional and experimental prognostic markers in prostate cancer: a pilot study. Br. J. Cancer 78, 246–250 (1998).
Tewari, A. & Narayan, P. Novel staging tool for localized prostate cancer: a pilot study using genetic adaptive neural networks. J. Urol. 160, 430–436 (1998).
Borque, A. et al. The use of neural networks and logistic regression analysis for predicting pathological stage in men undergoing radical prostatectomy: a population based study. J. Urol. 166, 1672–1678 (2001).
Veltri, R. W. et al. Comparison of logistic regression and neural net modeling for prediction of prostate cancer pathologic stage. Clin. Chem. 48, 1828–1834 (2002).
Zlotta, A. R. et al. An artificial neural network for prostate cancer staging when serum prostate specific antigen is 10 ng/ml or less. J. Urol. 169, 1724–1728 (2003).
Han, M., Snow, P. B., Brandt, J. M. & Partin, A. W. Evaluation of artificial neural networks for the prediction of pathologic stage in prostate carcinoma. Cancer 91, 1661–1666 (2001).
Gamito, E. J., Stone, N. N., Batuello, J. T. & Crawford, E. D. Use of artificial neural networks in the clinical staging of prostate cancer: implications for prostate brachytherapy. Tech. Urol. 6, 60–63 (2000).
Batuello, J. T. et al. Artificial neural network model for the assessment of lymph node spread in patients with clinically localized prostate cancer. Urology 57, 481–485 (2001).
Chakraborty, S., Ghosh, M., Maiti, T. & Tewari, A. Bayesian neural networks for bivariate binary data: an application to prostate cancer study. Stat. Med. 24, 3645–3662 (2005).
Regnier-Coudert, O. et al. Machine learning for improved pathological staging of prostate cancer: a performance comparison on a range of classifiers. Artif. Intell. Med. 55, 25–35 (2012).
Mattfeldt, T., Kestler, H. A., Hautmann, R. & Gottfried, H. W. Prediction of postoperative prostatic cancer stage on the basis of systematic biopsies using two types of artificial neural networks. Eur. Urol. 39, 530–536 (2001).
Matsui, Y. et al. Artificial neural network analysis for predicting pathological stage of clinically localized prostate cancer in the Japanese population. Jpn J. Clin. Oncol. 32, 530–535 (2002).
Walsh, P. C., Partin, A. W. & Epstein, J. I. Cancer control and quality of life following anatomical radical retropubic prostatectomy: results at 10 years. J. Urol. 152, 1831–1836 (1994).
Mattfeldt, T., Kestler, H. A., Hautmann, R. & Gottfried, H. W. Prediction of prostatic cancer progression after radical prostatectomy using artificial neural networks: a feasibility study. BJU Int. 84, 316–323 (1999).
Porter, C. et al. Artificial neural network model to predict biochemical failure after radical prostatectomy. Mol. Urol. 5, 159–162 (2001).
Han, M. et al. A neural network predicts progression for men with Gleason score 3+4 versus 4+3 tumors after radical prostatectomy. Urology 56, 994–999 (2000).
Potter, S. R. et al. Genetically engineered neural networks for predicting prostate cancer progression after radical prostatectomy. Urology 54, 791–795 (1999).
Ziada, A. M. et al. Impact of different variables on the outcome of patients with clinically confined prostate carcinoma: prediction of pathologic stage and biochemical failure using an artificial neural network. Cancer 91 (Suppl. 8), 1653–1660 (2001).
Tewari, A. et al. Genetic adaptive neural network to predict biochemical failure after radical prostatectomy: a multi-institutional study. Mol. Urol. 5, 163–169 (2001).
Shariat, S. F., Karakiewicz, P. I., Roehrborn, C. G. & Kattan, M. W. An updated catalog of prostate cancer predictive tools. Cancer 113, 3075–3099 (2008).
Abdel-Wahab, M. et al. ACR Appropriateness Criteria® external-beam radiation therapy treatment planning for clinically localized prostate cancer. J. Am. Coll. Radiol. 9, 233–238 (2012).
Wells, D. M. & Niederer, J. A medical expert system approach using artificial neural networks for standardized treatment planning. Int. J. Radiat. Oncol. Biol. Phys. 41, 173–182 (1998).
Lennernäs, B., Sandberg, D., Albertsson, P., Silén, A. & Isacsson, U. The effectiveness of artificial neural networks in evaluating treatment plans for patients requiring external beam radiotherapy. Oncol. Rep. 12, 1065–1070 (2004).
Gulliford, S. L., Webb, S., Rowbottom, C. G., Corne, D. W. & Dearnaley, D. P. Use of artificial neural networks to predict biological outcomes for patients receiving radical radiotherapy of the prostate. Radiother. Oncol. 71, 3–12 (2004).
Pella, A. et al. Use of machine learning methods for prediction of acute toxicity in organs at risk following prostate radiotherapy. Med. Phys. 38, 2859–2867 (2011).
Shariat, S. F., Karakiewicz, P. I., Margulis, V. & Kattan, M. W. Inventory of prostate cancer predictive tools. Curr. Opin. Urol. 18, 279–296 (2008).
Chiu, J. S. et al. Artificial neural network to predict skeletal metastasis in patients with prostate cancer. J. Med. Syst. 33, 91–100 (2009).
Jung, K., Miller, K., Wirth, M., Albrecht, M. & Lein, M. Bone turnover markers as predictors of mortality risk in prostate cancer patients with bone metastases following treatment with zoledronic acid. Eur. Urol. 59, 604–612 (2011).
Kawakami, S. et al. Development, validation, and head-to-head comparison of logistic regression-based nomograms and artificial neural network models predicting prostate cancer on initial extended biopsy. Eur. Urol. 54, 601–611 (2008).
Ecke, T. H. et al. External validation of an artificial neural network and two nomograms for prostate cancer detection. ISRN Urol. 2012, 643181 (2012).
Center, M. M. et al. International variation in prostate cancer incidence and mortality rates. Eur. Urol. 61, 1079–1092 (2012).
Finne, P. et al. 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 (2004).
Stephan, C. et al. Comparison of two different artificial neural networks for prostate biopsy indication in two different patient populations. Urology 70, 596–601 (2007).
Gosselaar, C., Kranse, R., Roobol, M. J., Roemeling, S. & Schroder, F. H. The interobserver variability of digital rectal examination in a large randomized trial for the screening of prostate cancer. Prostate 68, 985–993 (2008).
Pierorazio, P. M. et al. Preoperative characteristics of high-Gleason disease predictive of favourable pathological and clinical outcomes at radical prostatectomy. BJU Int. 110, 1122–1128 (2012).
Acknowledgements
The authors are supported by the Berliner Sparkassenstiftung Medizin and the Wilhelm Sander-Stiftung (grant 2010.111.1).
Author information
Authors and Affiliations
Contributions
X. Hu, H.-A. Meyer and C. Stephan researched the data for the article. H. Cammann, H.-A. Meyer, K. Jung and C. Stephan contributed to discussions of content. X. Hu, H. Cammann, K. Jung and C. Stephan wrote the manuscript, and X. Hu, K. Miller, K. Jung and C. Stephan edited the manuscript before submission.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Rights and permissions
About this article
Cite this article
Hu, X., Cammann, H., Meyer, HA. et al. Artificial neural networks and prostate cancer—tools for diagnosis and management. Nat Rev Urol 10, 174–182 (2013). https://doi.org/10.1038/nrurol.2013.9
Published:
Issue Date:
DOI: https://doi.org/10.1038/nrurol.2013.9
This article is cited by
-
Use of artificial intelligence in discerning the need for prostate biopsy and readiness for clinical practice: a systematic review protocol
Systematic Reviews (2023)
-
External validation of the computerized analysis of TRUS of the prostate with the ANNA/C-TRUS system: a potential role of artificial intelligence for improving prostate cancer detection
World Journal of Urology (2022)
-
Comorbidity and osteoporotic fracture: approach through predictive modeling techniques using the OSTEOMED registry
Aging Clinical and Experimental Research (2022)
-
Development and head-to-head comparison of machine-learning models to identify patients requiring prostate biopsy
BMC Urology (2021)
-
Artificial intelligence development for detecting prostate cancer in MRI
Egyptian Journal of Radiology and Nuclear Medicine (2021)