Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm: A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study

Research output: Contribution to journalArticlepeer-review

Standard

Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm : A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study. / Shouval, Roni; Labopin, Myriam; Bondi, Ori; Mishan-Shamay, Hila; Shimoni, Avichai; Ciceri, Fabio; Esteve, Jordi; Giebel, Sebastian; Gorin, Norbert C; Schmid, Christoph; Polge, Emmanuelle; Aljurf, Mahmoud; Kroger, Nicolaus; Craddock, Charles; Bacigalupo, Andrea; Cornelissen, Jan J; Baron, Frederic; Unger, Ron; Nagler, Arnon; Mohty, Mohamad.

In: Journal of Clinical Oncology , Vol. 33, No. 28, 01.10.2015, p. 3144-3151.

Research output: Contribution to journalArticlepeer-review

Harvard

Shouval, R, Labopin, M, Bondi, O, Mishan-Shamay, H, Shimoni, A, Ciceri, F, Esteve, J, Giebel, S, Gorin, NC, Schmid, C, Polge, E, Aljurf, M, Kroger, N, Craddock, C, Bacigalupo, A, Cornelissen, JJ, Baron, F, Unger, R, Nagler, A & Mohty, M 2015, 'Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm: A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study', Journal of Clinical Oncology , vol. 33, no. 28, pp. 3144-3151. https://doi.org/10.1200/JCO.2014.59.1339

APA

Shouval, R., Labopin, M., Bondi, O., Mishan-Shamay, H., Shimoni, A., Ciceri, F., Esteve, J., Giebel, S., Gorin, N. C., Schmid, C., Polge, E., Aljurf, M., Kroger, N., Craddock, C., Bacigalupo, A., Cornelissen, J. J., Baron, F., Unger, R., Nagler, A., & Mohty, M. (2015). Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm: A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study. Journal of Clinical Oncology , 33(28), 3144-3151. https://doi.org/10.1200/JCO.2014.59.1339

Vancouver

Author

Shouval, Roni ; Labopin, Myriam ; Bondi, Ori ; Mishan-Shamay, Hila ; Shimoni, Avichai ; Ciceri, Fabio ; Esteve, Jordi ; Giebel, Sebastian ; Gorin, Norbert C ; Schmid, Christoph ; Polge, Emmanuelle ; Aljurf, Mahmoud ; Kroger, Nicolaus ; Craddock, Charles ; Bacigalupo, Andrea ; Cornelissen, Jan J ; Baron, Frederic ; Unger, Ron ; Nagler, Arnon ; Mohty, Mohamad. / Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm : A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study. In: Journal of Clinical Oncology . 2015 ; Vol. 33, No. 28. pp. 3144-3151.

Bibtex

@article{014af4a5228d4bdbbf7bda2e55230503,
title = "Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm: A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study",
abstract = "PURPOSE: Allogeneic hematopoietic stem-cell transplantation (HSCT) is potentially curative for acute leukemia (AL), but carries considerable risk. Machine learning algorithms, which are part of the data mining (DM) approach, may serve for transplantation-related mortality risk prediction.PATIENTS AND METHODS: This work is a retrospective DM study on a cohort of 28,236 adult HSCT recipients from the AL registry of the European Group for Blood and Marrow Transplantation. The primary objective was prediction of overall mortality (OM) at 100 days after HSCT. Secondary objectives were estimation of nonrelapse mortality, leukemia-free survival, and overall survival at 2 years. Donor, recipient, and procedural characteristics were analyzed. The alternating decision tree machine learning algorithm was applied for model development on 70% of the data set and validated on the remaining data.RESULTS: OM prevalence at day 100 was 13.9% (n=3,936). Of the 20 variables considered, 10 were selected by the model for OM prediction, and several interactions were discovered. By using a logistic transformation function, the crude score was transformed into individual probabilities for 100-day OM (range, 3% to 68%). The model's discrimination for the primary objective performed better than the European Group for Blood and Marrow Transplantation score (area under the receiver operating characteristics curve, 0.701 v 0.646; P<.001). Calibration was excellent. Scores assigned were also predictive of secondary objectives.CONCLUSION: The alternating decision tree model provides a robust tool for risk evaluation of patients with AL before HSCT, and is available online (http://bioinfo.lnx.biu.ac.il/∼bondi/web1.html). It is presented as a continuous probabilistic score for the prediction of day 100 OM, extending prediction to 2 years. The DM method has proved useful for clinical prediction in HSCT.",
keywords = "Adult, Algorithms, Area Under Curve, Data Mining, Decision Support Techniques, Decision Trees, Disease Progression, Disease-Free Survival, Europe, Female, Hematopoietic Stem Cell Transplantation, Humans, Kaplan-Meier Estimate, Leukemia, Myeloid, Acute, Logistic Models, Machine Learning, Male, Middle Aged, Postoperative Complications, Precursor Cell Lymphoblastic Leukemia-Lymphoma, Predictive Value of Tests, ROC Curve, Registries, Reproducibility of Results, Retrospective Studies, Risk Assessment, Risk Factors, Time Factors, Transplantation, Homologous, Treatment Outcome, Journal Article, Research Support, Non-U.S. Gov't, Validation Studies",
author = "Roni Shouval and Myriam Labopin and Ori Bondi and Hila Mishan-Shamay and Avichai Shimoni and Fabio Ciceri and Jordi Esteve and Sebastian Giebel and Gorin, {Norbert C} and Christoph Schmid and Emmanuelle Polge and Mahmoud Aljurf and Nicolaus Kroger and Charles Craddock and Andrea Bacigalupo and Cornelissen, {Jan J} and Frederic Baron and Ron Unger and Arnon Nagler and Mohamad Mohty",
note = "{\textcopyright} 2015 by American Society of Clinical Oncology.",
year = "2015",
month = oct,
day = "1",
doi = "10.1200/JCO.2014.59.1339",
language = "English",
volume = "33",
pages = "3144--3151",
journal = "Journal of Clinical Oncology ",
issn = "0732-183X",
publisher = "American Society of Clinical Oncology",
number = "28",

}

RIS

TY - JOUR

T1 - Prediction of Allogeneic Hematopoietic Stem-Cell Transplantation Mortality 100 Days After Transplantation Using a Machine Learning Algorithm

T2 - A European Group for Blood and Marrow Transplantation Acute Leukemia Working Party Retrospective Data Mining Study

AU - Shouval, Roni

AU - Labopin, Myriam

AU - Bondi, Ori

AU - Mishan-Shamay, Hila

AU - Shimoni, Avichai

AU - Ciceri, Fabio

AU - Esteve, Jordi

AU - Giebel, Sebastian

AU - Gorin, Norbert C

AU - Schmid, Christoph

AU - Polge, Emmanuelle

AU - Aljurf, Mahmoud

AU - Kroger, Nicolaus

AU - Craddock, Charles

AU - Bacigalupo, Andrea

AU - Cornelissen, Jan J

AU - Baron, Frederic

AU - Unger, Ron

AU - Nagler, Arnon

AU - Mohty, Mohamad

N1 - © 2015 by American Society of Clinical Oncology.

PY - 2015/10/1

Y1 - 2015/10/1

N2 - PURPOSE: Allogeneic hematopoietic stem-cell transplantation (HSCT) is potentially curative for acute leukemia (AL), but carries considerable risk. Machine learning algorithms, which are part of the data mining (DM) approach, may serve for transplantation-related mortality risk prediction.PATIENTS AND METHODS: This work is a retrospective DM study on a cohort of 28,236 adult HSCT recipients from the AL registry of the European Group for Blood and Marrow Transplantation. The primary objective was prediction of overall mortality (OM) at 100 days after HSCT. Secondary objectives were estimation of nonrelapse mortality, leukemia-free survival, and overall survival at 2 years. Donor, recipient, and procedural characteristics were analyzed. The alternating decision tree machine learning algorithm was applied for model development on 70% of the data set and validated on the remaining data.RESULTS: OM prevalence at day 100 was 13.9% (n=3,936). Of the 20 variables considered, 10 were selected by the model for OM prediction, and several interactions were discovered. By using a logistic transformation function, the crude score was transformed into individual probabilities for 100-day OM (range, 3% to 68%). The model's discrimination for the primary objective performed better than the European Group for Blood and Marrow Transplantation score (area under the receiver operating characteristics curve, 0.701 v 0.646; P<.001). Calibration was excellent. Scores assigned were also predictive of secondary objectives.CONCLUSION: The alternating decision tree model provides a robust tool for risk evaluation of patients with AL before HSCT, and is available online (http://bioinfo.lnx.biu.ac.il/∼bondi/web1.html). It is presented as a continuous probabilistic score for the prediction of day 100 OM, extending prediction to 2 years. The DM method has proved useful for clinical prediction in HSCT.

AB - PURPOSE: Allogeneic hematopoietic stem-cell transplantation (HSCT) is potentially curative for acute leukemia (AL), but carries considerable risk. Machine learning algorithms, which are part of the data mining (DM) approach, may serve for transplantation-related mortality risk prediction.PATIENTS AND METHODS: This work is a retrospective DM study on a cohort of 28,236 adult HSCT recipients from the AL registry of the European Group for Blood and Marrow Transplantation. The primary objective was prediction of overall mortality (OM) at 100 days after HSCT. Secondary objectives were estimation of nonrelapse mortality, leukemia-free survival, and overall survival at 2 years. Donor, recipient, and procedural characteristics were analyzed. The alternating decision tree machine learning algorithm was applied for model development on 70% of the data set and validated on the remaining data.RESULTS: OM prevalence at day 100 was 13.9% (n=3,936). Of the 20 variables considered, 10 were selected by the model for OM prediction, and several interactions were discovered. By using a logistic transformation function, the crude score was transformed into individual probabilities for 100-day OM (range, 3% to 68%). The model's discrimination for the primary objective performed better than the European Group for Blood and Marrow Transplantation score (area under the receiver operating characteristics curve, 0.701 v 0.646; P<.001). Calibration was excellent. Scores assigned were also predictive of secondary objectives.CONCLUSION: The alternating decision tree model provides a robust tool for risk evaluation of patients with AL before HSCT, and is available online (http://bioinfo.lnx.biu.ac.il/∼bondi/web1.html). It is presented as a continuous probabilistic score for the prediction of day 100 OM, extending prediction to 2 years. The DM method has proved useful for clinical prediction in HSCT.

KW - Adult

KW - Algorithms

KW - Area Under Curve

KW - Data Mining

KW - Decision Support Techniques

KW - Decision Trees

KW - Disease Progression

KW - Disease-Free Survival

KW - Europe

KW - Female

KW - Hematopoietic Stem Cell Transplantation

KW - Humans

KW - Kaplan-Meier Estimate

KW - Leukemia, Myeloid, Acute

KW - Logistic Models

KW - Machine Learning

KW - Male

KW - Middle Aged

KW - Postoperative Complications

KW - Precursor Cell Lymphoblastic Leukemia-Lymphoma

KW - Predictive Value of Tests

KW - ROC Curve

KW - Registries

KW - Reproducibility of Results

KW - Retrospective Studies

KW - Risk Assessment

KW - Risk Factors

KW - Time Factors

KW - Transplantation, Homologous

KW - Treatment Outcome

KW - Journal Article

KW - Research Support, Non-U.S. Gov't

KW - Validation Studies

U2 - 10.1200/JCO.2014.59.1339

DO - 10.1200/JCO.2014.59.1339

M3 - Article

C2 - 26240227

VL - 33

SP - 3144

EP - 3151

JO - Journal of Clinical Oncology

JF - Journal of Clinical Oncology

SN - 0732-183X

IS - 28

ER -