Effective Automated Prediction of Vertebral Column Pathologies Based on Logistic Model Tree with SMOTE Preprocessing

Karabulut E. M., İBRİKÇİ T.

JOURNAL OF MEDICAL SYSTEMS, cilt.38, sa.5, 2014 (SCI-Expanded) identifier identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 38 Sayı: 5
  • Basım Tarihi: 2014
  • Doi Numarası: 10.1007/s10916-014-0050-0
  • Dergi Adı: JOURNAL OF MEDICAL SYSTEMS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Çukurova Üniversitesi Adresli: Evet

Özet

This study develops a logistic model tree based automation system based on for accurate recognition of types of vertebral column pathologies. Six biomechanical measures are used for this purpose: pelvic incidence, pelvic tilt, lumbar lordosis angle, sacral slope, pelvic radius and grade of spondylolisthesis. A two-phase classification model is employed in which the first step is preprocessing the data by use of Synthetic Minority Over-sampling Technique (SMOTE), and the second one is feeding the classifier Logistic Model Tree (LMT) with the preprocessed data. We have achieved an accuracy of 89.73 %, and 0.964 Area Under Curve (AUC) in computer based automatic detection of the pathology. This was validated via a 10-fold-cross-validation experiment conducted on clinical records of 310 patients. The study also presents a comparative analysis of the vertebral column data with the use of several machine learning algorithms.

This study develops a logistic model tree based automation system based on for accurate recognition of types of vertebral column pathologies. Six biomechanical measures are used for this purpose: pelvic incidence, pelvic tilt, lumbar lordosis angle, sacral slope, pelvic radius and grade of spondylolisthesis. A two-phase classification model is employed in which the first step is preprocessing the data by use of Synthetic Minority Over-sampling Technique (SMOTE), and the second one is feeding the classifier Logistic Model Tree (LMT) with the preprocessed data. We have achieved an accuracy of 89.73 %, and 0.964 Area Under Curve (AUC) in computer-based automatic detection of the pathology. This was validated via a 10-fold-cross-validation experiment conducted on clinical records of 310 patients. The study also presents a comparative analysis of the vertebral column data with the use of several machine learning algorithms.