Acute-on-chronic liver failure mortality prediction using an artificial neural network

Balaji Musunuri, Shiran Shetty, Dasharathraj K. Shetty, Manjunath K. Vanahalli, Aditya Pradhan, Nithesh Naik, Rahul Paul

Research output: Contribution to journalArticlepeer-review

18 Citations (Scopus)


Acute-on-chronic liver failure (ACLF) is a clinical syndrome affecting patients with chronic liver disease characterized by abrupt hepatic decompensation and associated with high short-term mortality. It is characterized by intense systemic inflammation, organ failure, and a poor prognosis. Using certain liver-specific prognostic scores, and organ failures, it is possible to triage and prognosticate the outcome of patients with ACLF. This work investigates the role of the artificial neural network (ANN), which functionally mimic biological neural systems, in predicting 90-day liver disease-related mortality. This study evaluated ANN among patients with ACLF. An accuracy of 94.12% was noticed at predicting 30-day mortality and 88.2% at predicting 90-day mortality, with an area under the curve of 0.915 and 0.921 respectively. ANN plays a very important role in predicting short term mortality patients with high accuracy. Its application in patients of ACLF is promising as it automates and eases the method of identifying those patients at higher risk of mortality. The application of ANN in this field has a vast potential for assisting clinicians in decision making, triaging of patients requiring emergent liver transplantation, and predicting mortality and complications.

Original languageEnglish
Pages (from-to)187-196
Number of pages10
JournalEngineered Science
Publication statusPublished - 2021

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Physical and Theoretical Chemistry
  • Chemistry (miscellaneous)
  • Materials Science(all)
  • Energy Engineering and Power Technology
  • Artificial Intelligence
  • Applied Mathematics


Dive into the research topics of 'Acute-on-chronic liver failure mortality prediction using an artificial neural network'. Together they form a unique fingerprint.

Cite this