Contribution of artificial neural networks to the classification and treatment of patients with uninvestigated dyspepsia

Angelo Andriulli, E. Grossi, M. Buscema, V. Festa, N. M. Intraligi, P. Dominici, R. Cerutti, F. Perri, S. Lupo, N. Casu, B. Chessa, G. Pomarico, G. Assenza, D. Iannuziello, G. Fuggi, P. Castagnoli, G. Dante, S. Bellentani, L. Vitale, U. Bigazzi & 36 others G. Giudici, G. Minoli, T. Luisi, R. Marin, V. Di Ciommo, F. Montanaro, B. Gaetano, S. Rosati, A. Maurano, F. Pietropaolo, V. Dambruoso, L. Tomarelli, R. Curto, C. Di Franco, L. Barresi, L. Marzio, R. Grossi, G. Sbolli, R. Marmo, R. Suriani, L. Mazzucca, L. D'Anna, A. Grassi, S. Monaco, G. Scarpulla, R. Clemente, M. Quitadamo, A. Carrato, S. Ghione, D. Cassine, A. Longhini, P. Borelli, C. Cortelezzi, A. Spadaccini, F. Biandrate, N. Belluardo

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Objectives. To verify whether symptoms reported by patients with uninvestigated dyspepsia might be helpful in either classifying functional from organic dyspepsia (1st experiment), or recognising which Helicobacter pylori infected patients may benefit from eradication therapy (2nd experiment). Methods. We compared the performance of artificial neural networks and linear discriminant analysis in two experiments on a database including socio-demographic features, past medical history, alarming symptoms, and symptoms at presentation of 860 patients with uninvestigated dyspepsia enrolled in a large observational multi-centre Italian study. Results. In the 1st experiment, the best prediction for organic disease was given by the Sine Net model (specificity of 87.6% with 13 patients misclassified) and the best prediction for functional dyspepsia by the FF Bp model (sensitivity of 83.4% with 56 patients misclassified). The highest global accuracy of linear discriminant analysis was 65.1%, with 150 patients misclassified. In the 2nd experiment, the highest predictive performance was provided by the SelfDASn model: all infected patients who became symptom-free after successful eradicating treatment were correctly classified, whereas nine errors were made in forecasting patients who did not benefit from such a therapy. The highest global performance of linear discriminant analysis was 53.2%, with 37 patients misclassified. Conclusions. In patients with uninvestigated dyspepsia, artificial neural networks might have potential for categorising those affected by either organic or functional dyspepsia, as well as for identifying all Helicobacter pylori infected dyspeptic patients who will benefit from eradication.

Original languageEnglish
Pages (from-to)222-231
Number of pages10
JournalDigestive and Liver Disease
Volume35
Issue number4
DOIs
Publication statusPublished - Apr 2003

Fingerprint

Dyspepsia
Discriminant Analysis
Therapeutics
Helicobacter pylori
Demography
Databases

Keywords

  • Artificial neural networks
  • Dyspepsia
  • Helicobacter pylori

ASJC Scopus subject areas

  • Gastroenterology

Cite this

Contribution of artificial neural networks to the classification and treatment of patients with uninvestigated dyspepsia. / Andriulli, Angelo; Grossi, E.; Buscema, M.; Festa, V.; Intraligi, N. M.; Dominici, P.; Cerutti, R.; Perri, F.; Lupo, S.; Casu, N.; Chessa, B.; Pomarico, G.; Assenza, G.; Iannuziello, D.; Fuggi, G.; Castagnoli, P.; Dante, G.; Bellentani, S.; Vitale, L.; Bigazzi, U.; Giudici, G.; Minoli, G.; Luisi, T.; Marin, R.; Di Ciommo, V.; Montanaro, F.; Gaetano, B.; Rosati, S.; Maurano, A.; Pietropaolo, F.; Dambruoso, V.; Tomarelli, L.; Curto, R.; Di Franco, C.; Barresi, L.; Marzio, L.; Grossi, R.; Sbolli, G.; Marmo, R.; Suriani, R.; Mazzucca, L.; D'Anna, L.; Grassi, A.; Monaco, S.; Scarpulla, G.; Clemente, R.; Quitadamo, M.; Carrato, A.; Ghione, S.; Cassine, D.; Longhini, A.; Borelli, P.; Cortelezzi, C.; Spadaccini, A.; Biandrate, F.; Belluardo, N.

In: Digestive and Liver Disease, Vol. 35, No. 4, 04.2003, p. 222-231.

Research output: Contribution to journalArticle

Andriulli, A, Grossi, E, Buscema, M, Festa, V, Intraligi, NM, Dominici, P, Cerutti, R, Perri, F, Lupo, S, Casu, N, Chessa, B, Pomarico, G, Assenza, G, Iannuziello, D, Fuggi, G, Castagnoli, P, Dante, G, Bellentani, S, Vitale, L, Bigazzi, U, Giudici, G, Minoli, G, Luisi, T, Marin, R, Di Ciommo, V, Montanaro, F, Gaetano, B, Rosati, S, Maurano, A, Pietropaolo, F, Dambruoso, V, Tomarelli, L, Curto, R, Di Franco, C, Barresi, L, Marzio, L, Grossi, R, Sbolli, G, Marmo, R, Suriani, R, Mazzucca, L, D'Anna, L, Grassi, A, Monaco, S, Scarpulla, G, Clemente, R, Quitadamo, M, Carrato, A, Ghione, S, Cassine, D, Longhini, A, Borelli, P, Cortelezzi, C, Spadaccini, A, Biandrate, F & Belluardo, N 2003, 'Contribution of artificial neural networks to the classification and treatment of patients with uninvestigated dyspepsia', Digestive and Liver Disease, vol. 35, no. 4, pp. 222-231. https://doi.org/10.1016/S1590-8658(03)00057-4
Andriulli, Angelo ; Grossi, E. ; Buscema, M. ; Festa, V. ; Intraligi, N. M. ; Dominici, P. ; Cerutti, R. ; Perri, F. ; Lupo, S. ; Casu, N. ; Chessa, B. ; Pomarico, G. ; Assenza, G. ; Iannuziello, D. ; Fuggi, G. ; Castagnoli, P. ; Dante, G. ; Bellentani, S. ; Vitale, L. ; Bigazzi, U. ; Giudici, G. ; Minoli, G. ; Luisi, T. ; Marin, R. ; Di Ciommo, V. ; Montanaro, F. ; Gaetano, B. ; Rosati, S. ; Maurano, A. ; Pietropaolo, F. ; Dambruoso, V. ; Tomarelli, L. ; Curto, R. ; Di Franco, C. ; Barresi, L. ; Marzio, L. ; Grossi, R. ; Sbolli, G. ; Marmo, R. ; Suriani, R. ; Mazzucca, L. ; D'Anna, L. ; Grassi, A. ; Monaco, S. ; Scarpulla, G. ; Clemente, R. ; Quitadamo, M. ; Carrato, A. ; Ghione, S. ; Cassine, D. ; Longhini, A. ; Borelli, P. ; Cortelezzi, C. ; Spadaccini, A. ; Biandrate, F. ; Belluardo, N. / Contribution of artificial neural networks to the classification and treatment of patients with uninvestigated dyspepsia. In: Digestive and Liver Disease. 2003 ; Vol. 35, No. 4. pp. 222-231.
@article{fd6f336205fb46bbbe6e1779e338a378,
title = "Contribution of artificial neural networks to the classification and treatment of patients with uninvestigated dyspepsia",
abstract = "Objectives. To verify whether symptoms reported by patients with uninvestigated dyspepsia might be helpful in either classifying functional from organic dyspepsia (1st experiment), or recognising which Helicobacter pylori infected patients may benefit from eradication therapy (2nd experiment). Methods. We compared the performance of artificial neural networks and linear discriminant analysis in two experiments on a database including socio-demographic features, past medical history, alarming symptoms, and symptoms at presentation of 860 patients with uninvestigated dyspepsia enrolled in a large observational multi-centre Italian study. Results. In the 1st experiment, the best prediction for organic disease was given by the Sine Net model (specificity of 87.6{\%} with 13 patients misclassified) and the best prediction for functional dyspepsia by the FF Bp model (sensitivity of 83.4{\%} with 56 patients misclassified). The highest global accuracy of linear discriminant analysis was 65.1{\%}, with 150 patients misclassified. In the 2nd experiment, the highest predictive performance was provided by the SelfDASn model: all infected patients who became symptom-free after successful eradicating treatment were correctly classified, whereas nine errors were made in forecasting patients who did not benefit from such a therapy. The highest global performance of linear discriminant analysis was 53.2{\%}, with 37 patients misclassified. Conclusions. In patients with uninvestigated dyspepsia, artificial neural networks might have potential for categorising those affected by either organic or functional dyspepsia, as well as for identifying all Helicobacter pylori infected dyspeptic patients who will benefit from eradication.",
keywords = "Artificial neural networks, Dyspepsia, Helicobacter pylori",
author = "Angelo Andriulli and E. Grossi and M. Buscema and V. Festa and Intraligi, {N. M.} and P. Dominici and R. Cerutti and F. Perri and S. Lupo and N. Casu and B. Chessa and G. Pomarico and G. Assenza and D. Iannuziello and G. Fuggi and P. Castagnoli and G. Dante and S. Bellentani and L. Vitale and U. Bigazzi and G. Giudici and G. Minoli and T. Luisi and R. Marin and {Di Ciommo}, V. and F. Montanaro and B. Gaetano and S. Rosati and A. Maurano and F. Pietropaolo and V. Dambruoso and L. Tomarelli and R. Curto and {Di Franco}, C. and L. Barresi and L. Marzio and R. Grossi and G. Sbolli and R. Marmo and R. Suriani and L. Mazzucca and L. D'Anna and A. Grassi and S. Monaco and G. Scarpulla and R. Clemente and M. Quitadamo and A. Carrato and S. Ghione and D. Cassine and A. Longhini and P. Borelli and C. Cortelezzi and A. Spadaccini and F. Biandrate and N. Belluardo",
year = "2003",
month = "4",
doi = "10.1016/S1590-8658(03)00057-4",
language = "English",
volume = "35",
pages = "222--231",
journal = "Digestive and Liver Disease",
issn = "1590-8658",
publisher = "Elsevier B.V.",
number = "4",

}

TY - JOUR

T1 - Contribution of artificial neural networks to the classification and treatment of patients with uninvestigated dyspepsia

AU - Andriulli, Angelo

AU - Grossi, E.

AU - Buscema, M.

AU - Festa, V.

AU - Intraligi, N. M.

AU - Dominici, P.

AU - Cerutti, R.

AU - Perri, F.

AU - Lupo, S.

AU - Casu, N.

AU - Chessa, B.

AU - Pomarico, G.

AU - Assenza, G.

AU - Iannuziello, D.

AU - Fuggi, G.

AU - Castagnoli, P.

AU - Dante, G.

AU - Bellentani, S.

AU - Vitale, L.

AU - Bigazzi, U.

AU - Giudici, G.

AU - Minoli, G.

AU - Luisi, T.

AU - Marin, R.

AU - Di Ciommo, V.

AU - Montanaro, F.

AU - Gaetano, B.

AU - Rosati, S.

AU - Maurano, A.

AU - Pietropaolo, F.

AU - Dambruoso, V.

AU - Tomarelli, L.

AU - Curto, R.

AU - Di Franco, C.

AU - Barresi, L.

AU - Marzio, L.

AU - Grossi, R.

AU - Sbolli, G.

AU - Marmo, R.

AU - Suriani, R.

AU - Mazzucca, L.

AU - D'Anna, L.

AU - Grassi, A.

AU - Monaco, S.

AU - Scarpulla, G.

AU - Clemente, R.

AU - Quitadamo, M.

AU - Carrato, A.

AU - Ghione, S.

AU - Cassine, D.

AU - Longhini, A.

AU - Borelli, P.

AU - Cortelezzi, C.

AU - Spadaccini, A.

AU - Biandrate, F.

AU - Belluardo, N.

PY - 2003/4

Y1 - 2003/4

N2 - Objectives. To verify whether symptoms reported by patients with uninvestigated dyspepsia might be helpful in either classifying functional from organic dyspepsia (1st experiment), or recognising which Helicobacter pylori infected patients may benefit from eradication therapy (2nd experiment). Methods. We compared the performance of artificial neural networks and linear discriminant analysis in two experiments on a database including socio-demographic features, past medical history, alarming symptoms, and symptoms at presentation of 860 patients with uninvestigated dyspepsia enrolled in a large observational multi-centre Italian study. Results. In the 1st experiment, the best prediction for organic disease was given by the Sine Net model (specificity of 87.6% with 13 patients misclassified) and the best prediction for functional dyspepsia by the FF Bp model (sensitivity of 83.4% with 56 patients misclassified). The highest global accuracy of linear discriminant analysis was 65.1%, with 150 patients misclassified. In the 2nd experiment, the highest predictive performance was provided by the SelfDASn model: all infected patients who became symptom-free after successful eradicating treatment were correctly classified, whereas nine errors were made in forecasting patients who did not benefit from such a therapy. The highest global performance of linear discriminant analysis was 53.2%, with 37 patients misclassified. Conclusions. In patients with uninvestigated dyspepsia, artificial neural networks might have potential for categorising those affected by either organic or functional dyspepsia, as well as for identifying all Helicobacter pylori infected dyspeptic patients who will benefit from eradication.

AB - Objectives. To verify whether symptoms reported by patients with uninvestigated dyspepsia might be helpful in either classifying functional from organic dyspepsia (1st experiment), or recognising which Helicobacter pylori infected patients may benefit from eradication therapy (2nd experiment). Methods. We compared the performance of artificial neural networks and linear discriminant analysis in two experiments on a database including socio-demographic features, past medical history, alarming symptoms, and symptoms at presentation of 860 patients with uninvestigated dyspepsia enrolled in a large observational multi-centre Italian study. Results. In the 1st experiment, the best prediction for organic disease was given by the Sine Net model (specificity of 87.6% with 13 patients misclassified) and the best prediction for functional dyspepsia by the FF Bp model (sensitivity of 83.4% with 56 patients misclassified). The highest global accuracy of linear discriminant analysis was 65.1%, with 150 patients misclassified. In the 2nd experiment, the highest predictive performance was provided by the SelfDASn model: all infected patients who became symptom-free after successful eradicating treatment were correctly classified, whereas nine errors were made in forecasting patients who did not benefit from such a therapy. The highest global performance of linear discriminant analysis was 53.2%, with 37 patients misclassified. Conclusions. In patients with uninvestigated dyspepsia, artificial neural networks might have potential for categorising those affected by either organic or functional dyspepsia, as well as for identifying all Helicobacter pylori infected dyspeptic patients who will benefit from eradication.

KW - Artificial neural networks

KW - Dyspepsia

KW - Helicobacter pylori

UR - http://www.scopus.com/inward/record.url?scp=0041306953&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041306953&partnerID=8YFLogxK

U2 - 10.1016/S1590-8658(03)00057-4

DO - 10.1016/S1590-8658(03)00057-4

M3 - Article

VL - 35

SP - 222

EP - 231

JO - Digestive and Liver Disease

JF - Digestive and Liver Disease

SN - 1590-8658

IS - 4

ER -