A machine learning approach to predict mortality due to immune-mediated thrombotic thrombocytopenic purpura

Aug 4, 2026

Uploaded by Aria

Mortality due to immune-mediated thrombotic thrombocytopenic purpura (iTTP) remains significant, and accurate risk prediction can support more individualized treatment decisions. The study externally evaluates the French Thrombotic Microangiopathy (TMA) Reference Score within the United States Thrombotic Microangiopathy (USTMA) iTTP database, then develops a new mortality prediction tool, the USTMA TTP Mortality Index, using gradient boosting machine learning.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Abstract

Background

Objectives

Methods

Results

Conclusion

[https://doi.org/10.1016/j.rpth.2024.102388](https://doi.org/10.1016/j.rpth.2024.102388) ORIGINAL ARTICLE A machine learning approach to predict mortality due to immune-mediated thrombotic thrombocytopenic purpura Mouhamed Yazan Abou-Ismail 1 | Chong Zhang2 | Angela P. Presson2 | Shruti Chaturvedi3 | Ana G. Antun4 | Andrew M. Farland5 | Ryan Woods5 | Ara Metjian6 | Yara A. Park7 | Gustaaf de Ridder7,8 | Briana Gibson7,9 | Raj S. Kasthuri 10 | Darla K. Liles 11 | Frank Akwaa 12 | Todd Clover 13 | Lisa Baumann Kreuziger14,15 | Meera Sridharan16 | Ronald S. Go16 | Keith R. McCrae17 | Harsh Vardhan Upreti3,18 | Radhika Gangaraju19 | Nicole K. Kocher19 | X. Long Zheng20,21 | Jay S. Raval22 | Camila Masias23 | Spero R. Cataland24 | Andrew D. Johnson25 | Elizabeth Davis26 | Michael D. Evans27 | Marshall Mazepa26 | Ming Y. Lim 1 | for the United States Thrombotic Microangiopathy Consortium 1Division of Hematology and Hematologic Malignancies, Department of Internal Medicine, University of Utah, Salt Lake City, Utah, USA 2Division of Epidemiology, Department of Internal Medicine, University of Utah, Salt Lake City, Utah, USA 3The Department of Medicine, Johns Hopkins University, Baltimore, Maryland, USA 4Department of Medicine, Emory University, Atlanta, Georgia, USA 5Department of Medicine, Wake Forest University, Winston-Salem, North Carolina, USA 6Department of Medicine, University of Colorado, Denver, Colorado, USA 7Department of Pathology and Laboratory Medicine, University of North Carolina, Chapel Hill, North Carolina, USA 8Geisinger Medical Laboratories, Danville, Pennsylvania, USA 9Department of Pathology and Laboratory Medicine, Emory University, Atlanta, Georgia, USA 10Department of Medicine, University of North Carolina, Chapel Hill, North Carolina, USA 11Department of Medicine, East Carolina University, Greenville, North Carolina, USA 12Department of Medicine, University of Rochester, Rochester, New York, USA 13St Charles Healthcare, Bend, Oregon, USA 14Versiti, Milwaukee, Wisconsin, USA 15Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin, USA 16Department of Medicine, Mayo Clinic, Rochester, Minnesota, USA 17Department of Medicine, Cleveland Clinic, Cleveland, Ohio, USA 18Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas, USA 19Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA 20Department of Pathology and Laboratory Medicine, University of Kansas Medical Center, Kansas City, Kansas, USA | Marshall Mazepa and Ming Y. Lim are cosenior authors. | | --- | | © 2024 The Author(s). Published by Elsevier Inc. on behalf of International Society on Thrombosis and Haemostasis. This is an open access article under the CC BY license ([http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/)). | Res Pract Thromb Haemost. 2024;8:e102388 [https://doi.org/10.1016/j.rpth.2024.102388](https://doi.org/10.1016/j.rpth.2024.102388) [www.rpthjournal.org](www.rpthjournal.org) - 1 of 11 2 of 11 - ABOU-ISMAIL ET AL. 21 Institute of Reproductive Medicine and Developmental Sciences, University of Kansas Medical Center, Kansas City, Kansas, USA 22Department of Pathology, University of New Mexico, Albuquerque, New Mexico, USA 23Baptist Health South Florida, Miami, Florida, USA 24Department of Medicine, The Ohio State University, Columbus, Ohio, USA 25Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota, USA 26Department of Medicine, University of Minnesota, Minneapolis, Minnesota, USA 27Clinical & Translational Science Institute, University of Minnesota, Minneapolis, Minnesota, USA Correspondence Mouhamed Yazan Abou-Ismail, Division of Hematology and Hematologic Malignancies, Department of Internal Medicine, 2000 Circle of Hope Drive, Salt Lake City, UT 84112, USA. Email: [yazan.abou-ismail@hsc.utah.e

FAQs

Why is predicting mortality in immune-mediated TTP important?

Mortality remains significant in iTTP, and predicting mortality risk can help individualize treatment decisions for patients.

What was the main objective of the study?

The study aimed to validate the French TMA Reference Score in the USTMA iTTP database and to develop a novel mortality prediction tool called the USTMA TTP Mortality Index.

How were the machine learning predictions developed?

Variables available at initial presentation (demographics, symptoms, laboratory findings) were used, and the model was built with gradient boosting machine learning implemented in the R package gbm.

Aug 4, 2026

Uploaded by Aria