Medical Risk Prediction Models: With Ties to Machine Learning

Medical Risk Prediction Models: With Ties to Machine Learning is a book that provides a hands-on approach for clinicians, epidemiologists, and professional statisticians to make or evaluate statistical prediction models based on data. It covers topics such as discrimination, calibration, and predictive performance, and offers R-code and illustrative examples to assist in interpretation of prediction performance. The book is authored by Thomas A. Gerds, a professor at the Biostatistics Unit at the University of Copenhagen, and Michael W. Kattan, Chair of the Department of Quantitative Health Sciences at Cleveland Clinic.

Format: Paperback / softback
Length: 290 pages
Publication date: 29 August 2022
Publisher: Taylor & Francis Ltd

Medical Risk Prediction Models: With Ties to Machine Learning is a comprehensive guide for clinicians, epidemiologists, and professional statisticians seeking to create or assess statistical prediction models based on data. The book delves into the individualized probability of a medical event within a specified time horizon, emphasizing the importance of understanding and interpreting these predictions. Authored by Thomas A. Gerds, a professor at the Biostatistics Unit at the University of Copenhagen and affiliated with the Danish Heart Foundation, and Michael W. Kattan, a highly cited author and Chair of the Department of Quantitative Health Sciences at Cleveland Clinic, the book offers a hands-on approach to making and evaluating statistical prediction models.

The authors employ a highly pedagogical style, avoiding mathematical notation, to explain the mathematical details involved in model creation and evaluation. This makes the book accessible to a wide range of readers, including those with limited mathematical backgrounds.

The book provides a comprehensive overview of the key concepts and techniques in medical risk prediction. It covers topics such as discrimination, calibration, and predictive performance with censored data and competing risks. Readers will learn how to correctly make online risk calculators from scratch and how to interpret the results of these calculations.

Furthermore, the book offers practical insights into the interpretation of prediction performance via benchmarks. It compares and combines rival modeling strategies using cross-validation, allowing readers to select the most appropriate modeling approach for their specific data and research objectives.

With its extensive coverage, comprehensive explanations, and practical examples, Medical Risk Prediction Models: With Ties to Machine Learning is an invaluable resource for clinicians, epidemiologists, and professional statisticians involved in medical risk prediction. It equips readers with the knowledge and skills necessary to create and evaluate accurate statistical prediction models, ultimately improving patient outcomes and healthcare decision-making.

Weight: 480g
Dimension: 156 x 233 x 20 (mm)
ISBN-13: 9780367673734