2019
DOI: 10.1182/bloodadvances.2019000613
|View full text |Cite
|
Sign up to set email alerts
|

Computational analysis of continuous body temperature provides early discrimination of graft-versus-host disease in mice

Abstract: Key Points Unsupervised machine learning analysis of continuous body temperature data revealed early signals of aGVHD in allo-HCT mice. Continuous measurement of body temperature is promising for early prediction of aGVHD in human allo-HCT patients.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
4
1

Relationship

2
3

Authors

Journals

citations
Cited by 5 publications
(3 citation statements)
references
References 13 publications
0
3
0
Order By: Relevance
“…This aim is exploratory; we expect to obtain pilot data to power a larger subsequent study to test correlations between wearable sensor data, symptoms data, and clinical outcomes. If sufficient data are available, we may also undertake a machine learning–based analysis, such as the one we recently described for the analysis of continuous temperature data for graft-versus-host disease prediction in an animal model [ 26 , 27 ].…”
Section: Methodsmentioning
confidence: 99%
“…This aim is exploratory; we expect to obtain pilot data to power a larger subsequent study to test correlations between wearable sensor data, symptoms data, and clinical outcomes. If sufficient data are available, we may also undertake a machine learning–based analysis, such as the one we recently described for the analysis of continuous temperature data for graft-versus-host disease prediction in an animal model [ 26 , 27 ].…”
Section: Methodsmentioning
confidence: 99%
“…Clustering is an example of unsupervised learning in which datapoints are grouped without prior knowledge of their labels, potentially identifying patient subgroups with similar risk profiles. This was utilized by Kuang et al to reveal differences in temperature between mice that developed acute GVHD post-HSCT and those that did not [25]. Reinforcement learning is a subset of ML inspired by behavioral psychology, where learning is achieved through trial and error.…”
Section: Machine Learning In Healthcarementioning
confidence: 99%
“…Pre-transplant donor and recipient clinical characteristics were used for the prediction model. Interestingly, using mouse model systems, Kuang et al [28] developed an early diagnostic model of acute GVHD. They analyzed continuous temperature profiles using principal component analysis and k-means clustering and captured temperature differences post-HSCT between mice that developed acute GVHD and those that did not.…”
Section: Post-hsct Complicationsmentioning
confidence: 99%