In this paper a new technique for classification of patients affected by Crohn’s disease (CD) is proposed. The proposed technique is based on a Kernel Support Vector Machine (KSVM) and it adopts a Stratified K-Fold Cross-Validation strategy to enhance the KSVM classifier reliability. Traditional manual classification methods require radiological expertise and they usually are very time-consuming. Accordingly to three expert radiologists, a dataset composed of 300 patients has been selected for KSVM training and validation. Each patient was codified by 22 extracted qualitative features and classified as Positive or Negative as the related histological specimen result showed the CD. The effectiveness of the proposed technique has been proved using a real human patient dataset collected at the University of Palermo Policlin-ico Hospital (UPPH dataset) and composed of 300 patients. The KSVM classification results have been compared against the histological specimen results, which are the adopted Ground-Truth for CD diagnosis. The achieved results (Sensitivity: 94,80%; Specificity: 100,00%; Negative Predictive Value: 95,06%; Precision: 100,00%; Accuracy: 97,40%; Error: 2,60%) show that the proposed technique results are comparable or even better than manual reference methods reported in literature.

Comelli, A., Terranova, M.C., Scopelliti, L., Salerno, S., Midiri, F., Lo Re, G., et al. (2018). A kernel support vector machine based technique for Crohnâ s disease classification in human patients. In L. Barolli and O. Terzo (a cura di), Complex, Intelligent, and Software Intensive Systems (pp. 262-273). Springer Verlag [10.1007/978-3-319-61566-0_25].

A kernel support vector machine based technique for Crohn’s disease classification in human patients

Comelli, Albert;Terranova, Maria Chiara;Scopelliti, Laura;Salerno, Sergio;Midiri, Federico;Lo Re, Giuseppe;Petrucci, Giovanni;Vitabile, Salvatore
2018-01-01

Abstract

In this paper a new technique for classification of patients affected by Crohn’s disease (CD) is proposed. The proposed technique is based on a Kernel Support Vector Machine (KSVM) and it adopts a Stratified K-Fold Cross-Validation strategy to enhance the KSVM classifier reliability. Traditional manual classification methods require radiological expertise and they usually are very time-consuming. Accordingly to three expert radiologists, a dataset composed of 300 patients has been selected for KSVM training and validation. Each patient was codified by 22 extracted qualitative features and classified as Positive or Negative as the related histological specimen result showed the CD. The effectiveness of the proposed technique has been proved using a real human patient dataset collected at the University of Palermo Policlin-ico Hospital (UPPH dataset) and composed of 300 patients. The KSVM classification results have been compared against the histological specimen results, which are the adopted Ground-Truth for CD diagnosis. The achieved results (Sensitivity: 94,80%; Specificity: 100,00%; Negative Predictive Value: 95,06%; Precision: 100,00%; Accuracy: 97,40%; Error: 2,60%) show that the proposed technique results are comparable or even better than manual reference methods reported in literature.
2018
Comelli, A., Terranova, M.C., Scopelliti, L., Salerno, S., Midiri, F., Lo Re, G., et al. (2018). A kernel support vector machine based technique for Crohnâ s disease classification in human patients. In L. Barolli and O. Terzo (a cura di), Complex, Intelligent, and Software Intensive Systems (pp. 262-273). Springer Verlag [10.1007/978-3-319-61566-0_25].
File in questo prodotto:
File Dimensione Formato  
10.1007-978-3-319-61566-0_25-main.pdf

Solo gestori archvio

Descrizione: Main PDF file
Dimensione 946.55 kB
Formato Adobe PDF
946.55 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10447/280201
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 16
  • ???jsp.display-item.citation.isi??? 12
social impact