Skip to main content
Springer Nature Link
Log in

Using Bayesian neural networks with ARD input selection to detect malignant ovarian masses prior to surgery

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

In this paper, we applied Bayesian multi-layer perceptrons (MLP) using the evidence procedure to predict malignancy of ovarian masses in a large (n = 1,066) multi-centre data set. Automatic relevance determination (ARD) was used to select the most relevant inputs. Fivefold cross-validation (5CV) and repeated 5CV was used to select the optimal combination of input set and number of hidden neurons. Results indicate good performance of the models with area under the receiver operating characteristic curve values of 0.93–0.94 on independent data. Comparison with a linear benchmark model and a previously developed logistic regression model shows that the present problem is very well linearly separable. A resampling analysis further shows that the number of hidden neurons specified in the ARD analyses for input selection may influence model performance. This paper shows that Bayesian MLPs, although not frequently used, are a useful tool for detecting malignant ovarian tumours.

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

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

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
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

References

  1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C, Thun MJ (2007) Cancer statistics, 2007. CA Cancer J Clin 57:43–66

    Article  Google Scholar 

  2. Vergote I, De Brabanter J, Fyles A, Bertelsen K, Einhorn N, Sevelda P, Gore ME, Karn J, Verrelst H, Sjovall K, Timmerman D, Vandewalle J, Van Gramberen M, Trope CG (2001) Prognostic factors in 1,545 patients with stage I invasive epithelial ovarium carcinoma: importance of degree of differential and cyst rupture in predicting relapse. Lancet 357:176–182

    Article  Google Scholar 

  3. Jacobs I, Oram D, Fairbanks J, Turner J, Frost C, Grudzinskas JG (1990) A risk of malignancy index incorporating CA 125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer. BJOG 97:922–929

    Article  Google Scholar 

  4. Tingulstad S, Hagen B, Skjeldestad FE, Onsrud M, Kiserud T, Halvorsen T, Nustad K (1996) Evaluation of a risk of malignancy index based on serum CA125, ultrasound findings and menopausal status in the pre-operative diagnosis of pelvic masses. BJOG 103:826–831

    Article  Google Scholar 

  5. Alcázar JL, Mercé LT, Laparte C, Jurado M, López-García G (2003) A new scoring system to differentiate benign from malignant adnexal masses. Am J Obstet Gynecol 188:685–692

    Article  Google Scholar 

  6. Sassone M, Timor-Tritsch IE, Artner A, Westhoff C, Warren W (1991) Transvaginal sonographic characterization of ovarian disease: evaluation of a new scoring system to predict ovarian malignancy. Obstet Gynecol 78:70–76

    Google Scholar 

  7. Lerner JP, Timor-Tritsch IE, Federman A, Abramovich G (1994) Transvaginal ultrasonographic characterization of ovarian masses with an improved, weighted scoring system. Am J Obstet Gynecol 170:81–85

    Google Scholar 

  8. Tailor A, Jurkovic D, Bourne TH, Collins WP, Campbell S (1997) Sonographic prediction of malignancy in adnexal masses using multivariate logistic regression analysis. Ultrasound Obstet Gynecol 10:41–47

    Article  Google Scholar 

  9. Timmerman D, Bourne TH, Tailor A, Collins WP, Verrelst H, Vandenberghe K, Vergote I (1999) A comparison of methods for preoperative discrimination between malignant and benign adnexal masses: the development of a new logistic regression model. Am J Obstet Gynecol 181:57–65

    Article  Google Scholar 

  10. Alcázar JL, Jurado M (1998) Using a logistic model to predict malignancy of adnexal masses based on menopausal status, ultrasound morphology, and color Doppler findings. Gynecol Oncol 69:146–150

    Article  Google Scholar 

  11. Tailor A, Jurkovic D, Bourne TH, Collins WP, Campbell S (1999) Sonographic prediction of malignancy in adnexal masses using an artificial neural network. BJOG 106:21–30

    Article  Google Scholar 

  12. Timmerman D, Verrelst H, Bourne TH, De Moor B, Collins WP, Vergote I, Vandewalle J (1999) Artificial neural network models for the preoperative discrimination between malignant and benign adnexal masses. Ultrasound Obstet Gynecol 13:17–25

    Article  Google Scholar 

  13. Biagiotti R, Desii C, Vanzi E, Gacci G (1999) Predicting ovarian malignancy: application of artificial neural networks to transvaginal and color Doppler flow US. Radiology 210:399–403

    Google Scholar 

  14. Clayton RD, Snowden S, Weston MJ, Mogensen O, Eastaugh J, Lane G (1999) Neural networks in the diagnosis of malignant ovarian tumours. BJOG 106:1078–1082

    Article  Google Scholar 

  15. Lu C, Van Gestel T, Suykens JAK, Van Huffel S, Vergote I, Timmerman D (2003) Preoperative prediction of malignancy of ovarian tumors using least squares support vector machines. Artif Intell Med 28:281–306

    Article  Google Scholar 

  16. Timmerman D (2000) Lack of standardization in gynecological ultrasonography (editorial). Ultrasound Obstet Gynecol 16:395–398

    Article  Google Scholar 

  17. Mol BWJ, Boll D, De Kanter M, Heintz APM, Sijmons EA, Oei SG, Bal H, Brölmann HAM (2001) Distinguishing the benign and malignant adnexal mass: an external validation of prognostic models. Gynecol Oncol 80:162–167

    Article  Google Scholar 

  18. Aslam N, Banerjee S, Carr JV, Savvas M, Hooper R, Jurkovic D (2000) Prospective evaluation of logistic regression models for the diagnosis of ovarian cancer. Obstet Gynecol 96:75–80

    Article  Google Scholar 

  19. Valentin L, Hagen B, Tingulstad S, Eik-Nes S (2001) Comparison of ‘pattern recognition’ and logistic regression models for discrimination between benign and malignant pelvic masses: a prospective cross validation. Ultrasound Obstet Gynecol 18:357–365

    Article  Google Scholar 

  20. Timmerman D, Valentin L, Bourne TH, Collins WP, Verrelst H, Vergote I (2000) Terms, definitions and measurements to describe the sonographic features of adnexal tumors: a consensus opinion from the International Ovarian Tumor Analysis (IOTA) group. Ultrasound Obstet Gynecol 16:500–505

    Article  Google Scholar 

  21. Timmerman D, Testa AC, Bourne T, Ferrazzi E, Ameye L, Konstantinovic ML, Van Calster B, Collins WP, Vergote I, Van Huffel S, Valentin L (2005) Logistic regression model to distinguish between the benign and malignant adnexal mass before surgery: a multicenter study by the International Ovarian Tumor Analysis Group. J Clin Oncol 23:8794–8801

    Article  Google Scholar 

  22. Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press, Oxford

    Google Scholar 

  23. MacKay DJC (1995) Probable networks and plausible predictions—a review of practical Bayesian methods for supervised neural networks. Netw-Comput Neural Syst 6:469–505

    Article  MATH  Google Scholar 

  24. MacKay DJC (1992) The evidence framework applied to classification networks. Neural Comput 4:698–714

    Google Scholar 

  25. Nabney IT, Evans DJ, Brulé Y, Gordon C (2005) Assessing the effectiveness of Bayesian feature selection. In: Husmeier D et al (eds) Probabilistic modelling in medical informatics and bioinformatics. Springer, London, pp 371–390

    Chapter  Google Scholar 

  26. Thodberg HH (1996) A review of Bayesian neural networks with an application to near infrared spectroscopy. IEEE Trans Neural Netw 7:56–72

    Article  Google Scholar 

  27. Berger JO (1985) Statistical design theory and Bayesian analysis. Springer, Berlin

    Google Scholar 

  28. Baum EB, Wilczek F (1988) Supervised learning of probability distributions by neural networks. In: Anderson D (ed) Advances in neural information processing systems, American Institute of Physics, New York, pp 52–61

    Google Scholar 

  29. Vellido A, Lisboa PJG (2001) An electronic commerce application of the Bayesian framework for MLPs: the effect of marginalisation and ARD. Neural Comput Appl 10:3–11

    Article  Google Scholar 

  30. Penny WD, Roberts SJ (1999) Bayesian neural networks for classification: how useful is the evidence framework? Neural Netw 12:877–892

    Article  Google Scholar 

  31. Marion G, Saad D (1995) A statistical mechanical analysis of a Bayesian inference scheme for an unrealizable rule. J Phys A-Math Gen 28:2159–2171

    Article  MATH  MathSciNet  Google Scholar 

  32. Neal RM (1996) Bayesian learning for neural networks. Springer, New York

    MATH  Google Scholar 

  33. Vivarelli F, Williams CKI (1997) Comparing Bayesian neural network algorithms for classifying segmented outdoor images. Neural Netw 14:427–437

    Article  Google Scholar 

  34. Lu C (2005) Probabilistic machine learning approaches to medical classification problems. PhD thesis, Department of Electrical Engineering, Katholieke Universiteit Leuven, Leuven

  35. Nabney IT (2002) NETLAB. Algorithms for pattern recognition. Springer, London

    MATH  Google Scholar 

  36. Kohavi R (1995) A study of cross-validation and bootstrap for accuracy estimation and model selection. In: Proceedings of the 14th international joint conference on artificial intelligence, Morgan Kaufman, San Francisco, pp 1137–1145

  37. Hanley JA, McNeil B (1982) The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29–36

    Google Scholar 

  38. Hastie TJ, Tibshirani RJ (1990) Generalized additive models. Chapman & Hall/CRC, Boca Raton

    MATH  Google Scholar 

  39. Dietterich TG (1998) Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput 10:1895–1923

    Article  Google Scholar 

  40. Kirk RE (2007) Effect magnitude: a different focus. J Stat Plan Inference 137:1634–1646

    Article  MATH  MathSciNet  Google Scholar 

  41. Bouckaert RR (2003) Choosing between two learning algorithms based on calibrated tests. In: Proceedings of the 20th international conference on machine learning, AAAI Press, Menlo Park, pp 51–58

  42. Nadeau I, Bengio Y (2000) Inference for the generalization error. In: Solla S et al (eds) Advances in neural information processing systems, vol 12, MIT Press, Cambridge, pp 307–313

    Google Scholar 

  43. Hand DJ (2006) Classifier technology and the illusion of progress. Stat Sci 21:1–15

    Article  MATH  MathSciNet  Google Scholar 

  44. Foresee FD, Hagan MT (1997) Gauss–Newton approximation to Bayesian learning. In proceedings of the 1997 international joint conference on neural networks, pp 1930–1935

  45. Etchells TA, Lisboa PJG (2006) Orthogonal search-based rule extraction (OSRE) for trained neural networks: a practical and efficient approach. IEEE Trans Neural Netw 17:374–384

    Article  Google Scholar 

Download references

Acknowledgments

This research was supported by the Research Council KUL: GOA-AMBioRICS, CoE EF/05/006 Optimization in Engineering, several PhD/postdoc and fellow grants, Flemish Government: FWO [PhD/postdoc grants, projects, G.0407.02 (support vector machines), G.0360.05 (EEG, Epileptic), G.0519.06 (Noninvasive brain oxygenation), FWO-G.0321.06 (tensors/spectral analysis), research communities (ICCoS, ANMMM)]; IWT (PhD grants), Belgian Federal Science Policy Office IUAP P5/22 (`Dynamical Systems and Control: Computation, Identification and Modelling’), EU: BIOPATTERN (FP6-2002-IST 508803), ETUMOUR (FP6-2002-LIFESCIHEALTH 503094), Healthagents (IST 2004–27214), and by research grants from the Swedish Medical Research Council (grants nos. K98-17X-11605-03A, K2001-72X-11605-06A and K2002-72X-11605-07B), by funds administered by Malmö University Hospital, Allmänna Sjukhusets i Malmö Stiftelse för bekämpande av cancer (the Malmö General Hospital Foundation for fighting against cancer) and ALF-medel (a Swedish governmental grant from the region of Scania).

Author information

Authors and Affiliations

  1. Department of Electrical Engineering (ESAT-SCD), Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, 3001, Leuven, Belgium

    Ben Van Calster & Sabine Van Huffel

  2. Department of Obstetrics and Gynaecology, University Hospitals Leuven, Herestraat 49, 3000, Leuven, Belgium

    Dirk Timmerman, Caroline Van Holsbeke & Ignace Vergote

  3. Neural Computing Research Group (NCRG), Aston University, Aston Triangle, Birmingham, B4 7ET, UK

    Ian T. Nabney

  4. Department of Obstetrics and Gynaecology, Malmö University Hospital, 205 02, Malmö, Sweden

    Lil Valentin

  5. Gynecologic Oncology Unit, Università Cattolica del Sacro Cuore, Largo Agostino Gemelli 8, 00168, Rome, Italy

    Antonia C. Testa

Authors
  1. Ben Van Calster
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Dirk Timmerman
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Ian T. Nabney
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Lil Valentin
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Antonia C. Testa
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Caroline Van Holsbeke
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Ignace Vergote
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Sabine Van Huffel
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Ben Van Calster.

Additional information

Parts of this work were originally presented at the 28th annual international conference of the IEEE Engineering in Medicine and Biology Society, 31 August–3 September 2006, New York City (NY), USA.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Van Calster, B., Timmerman, D., Nabney, I.T. et al. Using Bayesian neural networks with ARD input selection to detect malignant ovarian masses prior to surgery. Neural Comput & Applic 17, 489–500 (2008). https://doi.org/10.1007/s00521-007-0147-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-007-0147-1

Keywords