Predicting Death and Morbidity in Perforated Peptic Ulcer

Peptic ulcers are defined as defects in the gastrointestinal mucosa that extend through the muscularis mucosae. Although not being the most common complication, perforations stand out as being the complication with the highest mortality rate. To predict the probability of mortality, several scoring systems based on clinical and biochemical parameters, such as the Boey and PULP scoring system have been developed. This article explores, using data mining in the medical data available, how the scoring systems perform when trying to predict mortality and patients’ state complication. We also try to conclude, from the two scoring systems presented, which predicts better the situations described. Regarding the results, we concluded that the PULP scoring allows a better mortality prediction achieving, in this case, above 90% accuracy, however, the results may be inconclusive due to the lack of patients who died in the dataset used. Regarding the complications, we concluded that, on the other hand, the Boey system achieves better results leading to a better prediction when it comes to predicting patients’ state complication.

[1]  William W. Cohen Fast Effective Rule Induction , 1995, ICML.

[2]  J. Bendix,et al.  The Peptic Ulcer Perforation (PULP) score: a predictor of mortality following peptic ulcer perforation. A cohort study , 2012, Acta anaesthesiologica Scandinavica.

[3]  Clifford Goodman,et al.  The burden of selected digestive diseases in the United States. , 2002, Gastroenterology.

[4]  António Abelha,et al.  A Data Mining Approach for Cardiovascular Diagnosis , 2017, Open Comput. Sci..

[5]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[6]  Philip S. Yu,et al.  Top 10 algorithms in data mining , 2007, Knowledge and Information Systems.

[7]  V. Lohsiriwat,et al.  Perforated Peptic Ulcer: Clinical Presentation, Surgical Outcomes, and the Accuracy of the Boey Scoring System in Predicting Postoperative Morbidity and Mortality , 2008, World Journal of Surgery.

[8]  W. Najm Peptic ulcer disease. , 2011, Primary care.

[9]  Illhoi Yoo,et al.  Data Mining in Healthcare and Biomedicine: A Survey of the Literature , 2012, Journal of Medical Systems.

[10]  J. Richter,et al.  Trends and Outcomes of Hospitalizations for Peptic Ulcer Disease in the United States, 1993 to 2006 , 2010, Annals of surgery.

[11]  Krzysztof J. Cios,et al.  Uniqueness of medical data mining , 2002, Artif. Intell. Medicine.

[12]  Davis,et al.  Principles of Data Mining , 2001 .

[13]  António Abelha,et al.  Machine Learning in Nutritional Follow-up Research , 2017, Open Comput. Sci..

[14]  António Abelha,et al.  Knowledge discovery from surgical waiting lists , 2017 .

[15]  K. Søreide,et al.  Scoring systems for outcome prediction in patients with perforated peptic ulcer , 2013, Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine.

[16]  António Abelha,et al.  Predicting the need of Neonatal Resuscitation using Data Mining , 2017, EUSPN/ICTH.

[17]  N. N. Jani,et al.  Applying Naïve bayes, BayesNet, PART, JRip and OneR Algorithms on Hypothyroid Database for Comparative Analysis , 2014 .

[18]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[19]  J. Gisbert,et al.  Has H. pylori Prevalence in Bleeding Peptic Ulcer Been Underestimated? A Meta-Regression , 2011, The American Journal of Gastroenterology.

[20]  D. Irabor,et al.  An audit of peptic ulcer surgery in Ibadan, Nigeria. , 2006, West African journal of medicine.

[21]  Ron Kohavi,et al.  The Power of Decision Tables , 1995, ECML.