Ranking patients on the kidney transplant waiting list based on fuzzy inference system

Area of Science:

• Nephrology
• Artificial Intelligence
• Decision Support Systems

Background:

• Kidney transplantation is the optimal treatment for End-Stage Renal Disease (ESRD).
• Kidney allocation presents a significant challenge in transplantation.
• Developing an expert system to manage kidney allocation complexity is crucial.

Purpose of the Study:

• To develop an expert system, a Fuzzy Inference System (FIS), for ranking kidney transplant patients.
• To mimic expert decision-making in complex kidney allocation scenarios.

Main Methods:

• Identified key kidney allocation factors and weighted them using Intuitionistic Fuzzy Analytic Hierarchy Process (IF-AHP).
• Extracted fuzzy if-then rules from the United Network for Organ Sharing (UNOS) dataset using decision trees.
• Developed a Multi-Input Single-Output (MISO) Mamdani fuzzy inference system (FISKA) for patient ranking.

Main Results:

• The developed Fuzzy Inference System for Kidney Allocation (FISKA) was evaluated against point scoring and filtering systems.
• FISKA demonstrated greater acceptability among experts compared to conventional kidney allocation methods.

Conclusions:

• FISKA offers a valuable tool for enhancing kidney allocation systems, addressing the scarcity of donated organs.
• The proposed model is adaptable for improving allocation systems for other organs like lungs, liver, and heart.