A Systematic Review of Machine Learning Techniques in Hematopoietic Stem Cell Transplantation (HSCT)

Abstract

Machine learning techniques are widely used nowadays in the healthcare domain for the diagnosis, prognosis, and treatment of diseases. These techniques have applications in the field of hematopoietic cell transplantation (HCT), which is a potentially curative therapy for hematological malignancies. Herein, a systematic review of the application of machine learning (ML) techniques in the HCT setting was conducted. We examined the type of data streams included, specific ML techniques used, and type of clinical outcomes measured. A systematic review of English articles using PubMed, Scopus, Web of Science, and IEEE Xplore databases was performed. Search terms included “hematopoietic cell transplantation (HCT),” “autologous HCT,” “allogeneic HCT,” “machine learning,” and “artificial intelligence.” Only full-text studies reported between January 2015 and July 2020 were included. Data were extracted by two authors using predefined data fields. Following PRISMA guidelines, a total of 242 studies were identified, of which 27 studies met the inclusion criteria. These studies were sub-categorized into three broad topics and the type of ML techniques used included ensemble learning (63%), regression (44%), Bayesian learning (30%), and support vector machine (30%). The majority of studies examined models to predict HCT outcomes (e.g., survival, relapse, graft-versus-host disease). Clinical and genetic data were the most commonly used predictors in the modeling process. Overall, this review provided a systematic review of ML techniques applied in the context of HCT. The evidence is not sufficiently robust to determine the optimal ML technique to use in the HCT setting and/or what minimal data variables are required.

Monitoring Beliefs and Physiological Measures Using Wearable Sensors and Smartphone Technology Among Students at Risk of COVID-19: Protocol for a mHealth Study

“…We aim to use computational techniques to assess the relationship among self-reported symptom data, continuous heart rate data, and the incidence of clinical respiratory illness, which may be diagnosed as COVID-19 or other types of infections. We will build on analytic approaches already developed in our study on oncology patients (ie, patients undergoing hematopoietic stem cell transplantation, who develop fevers leading to changes in heart rate, step, and sleep owing to infections, graft-versus-host disease, or other causes; patients receiving chimeric antigen receptor T cells, who develop cytokine release syndrome, display some similarity to the pathophysiology of severe COVID-19) [ 24 , 25 ].…”

The Contemporary Approach to CALR-Positive Myeloproliferative Neoplasms

“…Machine learning is an application of AI that provides the capability to automatically learn and improve from experience without being explicitly programmed [ 119 ]. The process is an extension of statistical methods leading to predictions and automatic identification of patterns ending in performing tasks beyond human capabilities [ 120 ].…”

Dynamic forecasting of severe acute graft-versus-host disease after transplantation

“…Limited data capture, non-uniform sampling rates, and data integration issues have all been cited as primary challenges in applying ML in HSCT. 19 One recent study applied penalized logistic regression to vital signs (temperature, heart rate, etc.) that were consistently and frequently recorded within the first 10 days after HSCT 20 , and we would like to investigate if we could utilize additional evidence from other dynamic features that were more irregularly measured.…”