ATTEMPTING TO INCREASE THE CLINICAL DIAGNOSTIC RATE OF ACUTE INTESTINAL ISCHEMIA USING MACHINE LEARNING ALGORITHMS

  • Pîrvu Cătălin Alexandru Pius Brânzeu County Emergency Clinical Hospital Timisoara, Romania & Victor Babeș Medicine and Pharmacy University Timisoara, Romania
  • Cristian Nica Pius Brânzeu County Emergency Clinical Hospital Timisoara, Romania & Victor Babeș Medicine and Pharmacy University Timisoara, Romania
  • Mărgăritescu Dragoș Clinical County Emegrency Hospital of Craiova, Romania & University of Medicine and Pharmacy Craiova, Romania
  • Pătrașcu Ștefan Clinical County Emegrency Hospital of Craiova, Romania & University of Medicine and Pharmacy Craiova, Romania
  • Valeriu Șurlin Clinical County Emegrency Hospital of Craiova, Romania & University of Medicine and Pharmacy Craiova, Romania
  • Konstantions Sapalidis 3rd Department of Surgery, “AHEPA” University Hospital, Thessaloniki, Greece & Aristotle University of Thessaloniki, Medical School, Thessaloniki, Greece
  • Eugen Georgescu Clinical County Emegrency Hospital of Craiova, Romania & University of Medicine and Pharmacy Craiova, Romania
  • Ion Georgescu Clinical County Emegrency Hospital of Craiova, Romania & University of Medicine and Pharmacy Craiova, Romania
  • Stelian Pantea Pius Brânzeu County Emergency Clinical Hospital Timisoara, Romania & Victor Babeș Medicine and Pharmacy University Timisoara, Romania
Keywords: bowel infarction, acute abdomen, intestinal ischemia, machine learning

Abstract

Acute intestinal ischemia (AMI) is a life-threatening surgical emergency where more than half of the affected patients do not survive. In spite of the medical advance, mortal-ity rates remain high due to late diagnosis, when proper surgical management and reperfusion techniques do not conclude to a successful outcome.  The current study aims  to find a proper diagnosis method with a high-reliability rate using machine learning (ML) algorithms. Methods: In this prospective cross-sectional study, we have collected and evaluated over the course of two years a total of 147 patients with a clini-cal presentation resembling acute mesenteric ischemia. Five ML algorithms, including Random Forest, Logistic Regression, Gradient Boosted Trees, Naive Bayes, and Multi-ple Layer Perceptron, were compared for their reliability in diagnosing acute intestinal ischemia by using regular blood tests performed in the emergency room (ER), on top of the main clinical characteristics of the researched condition. An algorithm score using Gradient Boosted Trees and Logistic Regression proved good diagnostic performance with an AUROC 0.784, p<0.001, with a sensitivity of 83.8%, specificity of 58.2%, 70.5% positive predictive value, and 75% negative predictive value. The ML algorithm is use-ful in detecting AMI using only anamnesis data and regular laboratory blood tests available in the ER, although it was not internally validated.

References

[1] Mastoraki A., Mesenteric ischemia: Pathogenesis and challenging di-agnostic and therapeutic modalities. World Journal of Gastrointestinal Pathophysiolo-gy, 7(1), 125, 2016
[2]Bala M., Kashuk, Jeffry, Moore, Ernest E.; Kluger, Yoram; Biffl, Walter; Gomes, Carlos Augusto; Ben-Ishay, Offir; Rubinstein, Chen; Balogh, Zsolt J.; Civil, Ian; Cocco-lini, Federico; Leppaniemi, Ari; Peitzman, Andrew; Ansaloni, Luca; Sugrue, Michael; Sartelli, Massimo; Di Saverio, Salomone; Fraga, Gustavo P.; Catena, Fausto (2017). Acute mesenteric ischemia: guidelines of the World Society of Emergency Surgery. World Journal of Emergency Surgery, 12(1), 38.
[3] Florim, S.; Almeida, A.; Rocha, D.; Portugal, P. (2018). Acute mesenteric ischaemia: a pictorial review. Insights into Imaging.
[4]Patel, Amit; Kaleya, Ronald N.; Sammartano, Robert J. (1992). Pathophysiology of Mesenteric Ischemia. Surgical Clinics of North America, 72(1), 31–41.
[5]Stefan Acosta (2010). Epidemiology of Mesenteric Vascular Disease: Clinical Implica-tions. 23(1), 4–8.
[6]Jaclyn F. Hechtman; Patrick A. Lento; Irini Scordi-bello (2011). Ischemic bowel due to embolization from an isolated mobile thrombus of the ascending aorta: a case report and review of the literature. 32(2), 238–241.
[7]Hmoud, Bashar; Singal, Ashwani K.; Kamath, Patrick S. (2014). Mesenteric Venous Thrombosis. Journal of Clinical and Experimental Hepatology, 4(3), 257–263.
[8]Pasupathy, Sivabaskari; Tavella, Rosanna; McRae, Simon; Beltrame, John F (2015). Myocardial Infarction With Non-obstructive Coronary Arteries — Diagnosis and Management. European Cardiology Review, 10(2), 79
[9]Singh, Hardeep; Schiff, Gordon D; Graber, Mark L; Onakpoya, Igho; Thompson, Mat-thew J (2016). The global burden of diagnostic errors in primary care. BMJ Quality & Safety, (), bmjqs-2016-005401.
[10]Yu, Kun-Hsing; Beam, Andrew L.; Kohane, Isaac S. (2018). Artificial intelligence in healthcare. Nature Biomedical Engineering, 2(10), 719–731.
[11]Richens, Jonathan G.; Lee, Ciarán M.; Johri, Saurabh (2020). Improving the accura-cy of medical diagnosis with causal machine learning. Nature Communications, 11(1), 3923–.
[12]Lynam, Anita L.; Dennis, John M.; Owen, Katharine R.; Oram, Richard A.; Jones, An-gus G.; Shields, Beverley M.; Ferrat, Lauric A. (2020). Logistic regression has similar performance to optimised machine learning algorithms in a clinical setting: applica-tion to the discrimination between type 1 and type 2 diabetes in young adults. Diag-nostic and Prognostic Research, 4(1), 6
[13]Strand-Amundsen, Runar James; Tronstad, Christian; Reims, Henrik Mikael; Reinholt, Finn P; Hogetveit, Jan Olav; Tønnessen, Tor Inge (2018). Machine learning for in-traoperative prediction of viability in ischemic small intestine. Physiological Measure-ment.
[14]Zhuang, X., Chen, F., Zhou, Q. et al. A rapid preliminary prediction model for intesti-nal necrosis in acute mesenteric ischemia: a retrospective study. BMC Gastroenterol 21, 154 (2021).
[15]Hwang, Eui Jin; Park, Sunggyun; Jin, Kwang-Nam; Kim, Jung Im; Choi, So Young; Lee, Jong Hyuk; Goo, Jin Mo; Aum, Jaehong; Yim, Jae-Joon; Cohen, Julien G.; Ferretti, Gilbert R.; Park, Chang Min (2019). Development and Validation of a Deep Learn-ing–Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Ra-diographs. JAMA Network Open, 2(3), e191095.
[16]Ewout W Steyerberg; Sacha E Bleeker; Henriëtte A Moll; Diederick E Grobbee; Karel G.M Moons (2003). Internal and external validation of predictive models: A simula-tion study of bias and precision in small samples. , 56(5), 0–447.
Published
2021-07-09
How to Cite
[1]
P. Alexandru, “ATTEMPTING TO INCREASE THE CLINICAL DIAGNOSTIC RATE OF ACUTE INTESTINAL ISCHEMIA USING MACHINE LEARNING ALGORITHMS”, JSS, vol. 8, no. 2, Jul. 2021.