Use of Predictive Models to Determine Transplant Eligibility.

Area of Science:

• Medical Informatics
• Biostatistics
• Transplant Surgery

Background:

• Predictive modeling is crucial for estimating patient health outcomes.
• Current transplant eligibility models primarily focus on pre- or post-transplant survival.
• A gap exists in models that comprehensively integrate all transplant evaluation data.

Purpose of the Study:

• To summarize predictive models for transplant eligibility developed in the last five years.
• To highlight the application of novel data sources and methodological approaches in this field.
• To identify the need for composite predictions of post-transplant success.

Main Methods:

• Review of recent literature on predictive modeling in transplant eligibility.
• Focus on studies utilizing novel data sources and advanced methodologies.
• Analysis of existing approaches to transplant outcome prediction.

Main Results:

• The majority of recent models concentrate on patient survival.
• No existing studies have developed a composite prediction of post-transplant success using all evaluation features.
• Challenges remain due to the lack of curated, well-defined data for transplant candidates.

Conclusions:

• There is a need for more integrated predictive models in transplant eligibility.
• Advancements in data science offer potential for improved transplant outcome prediction.
• Future research should focus on creating comprehensive models using diverse data sources.