Knowledge discovery for pancreatic cancer using inductive logic programming. Issue 4 (1st August 2014)

• Record Type: Journal Article
• Authors: Qiu, Yushan; Shimada, Kazuaki; Hiraoka, Nobuyoshi; Maeshiro, Kensei; Ching, Wai‐Ki; Aoki‐Kinoshita, Kiyoko F.; Furuta, Koh
• Is Part Of: IET systems biology. Volume 8:Issue 4(2014)
• Page Start: 162
• Page End: 168
• Publication Date: 2014-08-01
• Subjects: cancer -- data mining -- inductive logic programming -- medical information systems -- patient treatment -- tumours; pancreatic cancer -- patient status -- inductive logic programming method -- clinical laboratory data -- disease characteristic prediction -- therapeutic modality selection -- disease knowledge discovery -- tumour differentiation -- lymph node metastasis status -- ILP model -- therapeutic strategy decision -- classification techniques; Systems biology -- Periodicals; Cell physiology -- Periodicals; Biological systems -- Mathematical models -- Periodicals; Genetics -- Mathematical models -- Periodicals; Computational biology -- Periodicals; 573
• DOI: 10.1049/iet-syb.2013.0044
• Languages: English
• ISSNs: 1751-8849

Abstract

Abstract : Pancreatic cancer is a devastating disease and predicting the status of the patients becomes an important and urgent issue. The authors explore the applicability of inductive logic programming (ILP) method in the disease and show that the accumulated clinical laboratory data can be used to predict disease characteristics, and this will contribute to the selection of therapeutic modalities of pancreatic cancer. The availability of a large amount of clinical laboratory data provides clues to aid in the knowledge discovery of diseases. In predicting the differentiation of tumour and the status of lymph node metastasis in pancreatic cancer, using the ILP model, three rules are developed that are consistent with descriptions in the literature. The rules that are identified are useful to detect the differentiation of tumour and the status of lymph node metastasis in pancreatic cancer and therefore contributed significantly to the decision of therapeutic strategies. In addition, the proposed method is compared with the other typical classification techniques and the results further confirm the superiority and merit of the proposed method.