Statistical Learning Approaches In Diagnosing Patients With Nontraumatic Acute Abdomen

A quick evaluation is required for patients with acute abdominal pain. It is crucial to differentiate between surgical and nonsurgical pathology. Practical and accurate tests are essential in this differentiation. Lately, D-dimer level has been found to be an important adjuvant in this diagnosis and obviously outperforms leukocyte count, which is widely used for diagnosis of certain cases. Here, we handle this problem from a statistical perspective and combine the information from leukocyte count with D-dimer level to increase the diagnostic accuracy of nontraumatic acute abdomen. For this purpose, various statistical learning algorithms are considered and model performances are assessed using several measures. Our results revealed that the näıve Bayes algorithm, robust quadratic discriminant analysis, bagged and boosted support vector machines, and single and bagged k-nearest neighbors provide an increase in diagnostic accuracies of up to 8.93% and 17.86% compared with D-dimer level and leukocyte count, respectively. Highest accuracy was obtained as 78.57% with the näıve Bayes algorithm. Analysis has been done via the R programming language based on the codes developed by the authors. A user-friendly web-tool is also developed to assist physicians in their decisions to differentially diagnose patients with acute abdomen. It is available at http://www.biosoft.hacettepe.edu.tr/DDNAA/.

[1]  Marcelo Dias,et al.  Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry. , 2013, Arquivos brasileiros de oftalmologia.

[2]  Namik Yilmaz,et al.  The value of D-dimer test in the diagnosis of patients with nontraumatic acute abdomen. , 2010, Ulusal travma ve acil cerrahi dergisi = Turkish journal of trauma & emergency surgery : TJTES.

[3]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[4]  Johan A. K. Suykens,et al.  Systematic benchmarking of microarray data classification: assessing the role of non-linearity and dimensionality reduction , 2004, Bioinform..

[5]  D Frager,et al.  CT of small-bowel obstruction: value in establishing the diagnosis and determining the degree and cause. , 1994, AJR. American journal of roentgenology.

[6]  Jaap Stoker,et al.  Imaging patients with acute abdominal pain. , 2009, Radiology.

[7]  Edvaldo Fahel,et al.  Non-Traumatic Acute Abdomen: Videolaparoscopic Approach , 1999, JSLS : Journal of the Society of Laparoendoscopic Surgeons.

[8]  B. Marincek,et al.  Nontraumatic abdominal emergencies: acute abdominal pain: diagnostic strategies , 2002, European Radiology.

[9]  Tong Wu,et al.  Diagnosis of colorectal cancer by near-infrared optical fiber spectroscopy and random forest. , 2015, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[10]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[11]  Thomas G. Dietterich.  Multiple Classifier Systems , 2000, Lecture Notes in Computer Science.

[12]  M. Pazzani,et al.  Improving Dementia Screening Tests with Machine Learning Methods , 1996 .

[13]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[14]  M. Bardella,et al.  IgA antigliadin antibodies, cellobiose/mannitol sugar test, and carotenemia in the diagnosis of and screening for celiac disease. , 1991, The American journal of gastroenterology.

[15]  S. Cartwright,et al.  Evaluation of acute abdominal pain in adults. , 2008, American family physician.

[16]  Shekhar Gogna,et al.  Comparative Analysis of Clinical, Radiological and Operative Findings in Acute Abdomen , 2015 .

[17]  Eric Goldberg,et al.  Acute abdominal pain. , 2006, The Medical clinics of North America.

[18]  R. Marshall,et al.  The Predictive Value of Simple Rules for Combining Two Diagnostic Tests , 1989 .

[19]  Max Kuhn,et al.  Building Predictive Models in R Using the caret Package , 2008 .

[20]  Alper Akcan,et al.  D‐dimer as a predictor of the need for laparotomy in patients with unclear non‐traumatic acute abdomen. A preliminary study , 2008, Scandinavian journal of clinical and laboratory investigation.

[21]  Vipin Kumar,et al.  Introduction to Data Mining, (First Edition) , 2005 .

[22]  Cengiz Sertkaya,et al.  Comparison of different methods for determining diabetes , 2014 .

[23]  Peter Filzmoser,et al.  An Object-Oriented Framework for Robust Multivariate Analysis , 2009 .

[24]  K. Hornik,et al.  Unbiased Recursive Partitioning: A Conditional Inference Framework , 2006 .

[25]  F. Kelvin,et al.  Obstruction of the small intestine: accuracy and role of CT in diagnosis. , 1993, Radiology.