Prognosis models for nasopharyngeal carcinoma recurrences by using tabu search algorithm

Area of Science:

• Oncology and Artificial Intelligence
• Computational Intelligence in Healthcare
• Data Mining and Predictive Analytics

Background:

• Recurrent nasopharyngeal carcinoma (NPC) poses significant mortality risks, necessitating improved treatment strategies.
• Metaheuristic algorithms (MH) and data mining are increasingly vital in healthcare for diagnostics and prediction.
• Existing prognoses for NPC recurrence require enhancement to improve patient survival and quality of life.

Purpose of the Study:

• To develop an Artificial Advisory Healthcare System (AAHS) for predicting NPC recurrence.
• To enhance treatment regimens and patient survival through accurate recurrence prediction.
• To leverage artificial intelligence for improved understanding and management of NPC.

Main Methods:

• Utilized the Tabu Search (TS) algorithm, enhanced with Dynamic Neighborhood Structure (DNHS), for data mining challenges.
• Developed three predictive models using patient data, incorporating increasing features at different treatment stages.
• Integrated these models into an AAHS for real-time recurrence prediction and treatment adjustment.

Main Results:

• The proposed Tabu Search Classifier Method (TSCM) accurately predicts NPC recurrence at various treatment stages.
• The three predictive models demonstrated superior performance compared to existing NPC recurrence prognoses.
• The AAHS facilitates timely adjustments to treatment plans based on predicted recurrence risk.

Conclusions:

• The developed AAHS effectively predicts NPC recurrence, aiding oncologists in proactive patient management.
• This AI-driven approach offers a significant advancement in predicting and preventing NPC recurrence.
• The study highlights the potential of metaheuristic algorithms and data mining in personalized cancer care.