% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Aldraimli:180190,
author = {M. Aldraimli and S. Osman and D. Grishchuck and S. Ingram
and R. Lyon and A. Mistry and J. Oliveira and R. Samuel and
L. E. A. Shelley and D. Soria and M. V. Dwek and M. E.
Aguado-Barrera and D. Azria and J. Chang-Claude$^*$ and A.
Dunning and A. Giraldo and S. Green and S.
Gutiérrez-Enríquez and C. Herskind and H. van Hulle and M.
Lambrecht and L. Lozza and T. Rancati and V. Reyes and B. S.
Rosenstein and D. de Ruysscher and M. C. de Santis and P.
Seibold$^*$ and E. Sperk and R. P. Symonds and H. Stobart
and B. Taboada-Valadares and C. J. Talbot and V. J. L.
Vakaet and A. Vega and L. Veldeman and M. R. Veldwijk and A.
Webb and C. Weltens and C. M. West and T. J. Chaussalet and
T. Rattay},
collaboration = {R. consortium},
title = {{D}evelopment and {O}ptimization of a {M}achine-{L}earning
{P}rediction {M}odel for {A}cute {D}esquamation {A}fter
{B}reast {R}adiation {T}herapy in the {M}ulticenter
{REQUITE} {C}ohort.},
journal = {Advances in radiation oncology},
volume = {7},
number = {3},
issn = {2452-1094},
address = {Amsterdam},
publisher = {Elsevier},
reportid = {DKFZ-2022-01162},
pages = {100890},
year = {2022},
abstract = {Some patients with breast cancer treated by surgery and
radiation therapy experience clinically significant
toxicity, which may adversely affect cosmesis and quality of
life. There is a paucity of validated clinical prediction
models for radiation toxicity. We used machine learning (ML)
algorithms to develop and optimise a clinical prediction
model for acute breast desquamation after whole breast
external beam radiation therapy in the prospective
multicenter REQUITE cohort study.Using demographic and
treatment-related features (m = 122) from patients (n =
2058) at 26 centers, we trained 8 ML algorithms with 10-fold
cross-validation in a 50:50 random-split data set with class
stratification to predict acute breast desquamation. Based
on performance in the validation data set, the logistic
model tree, random forest, and naïve Bayes models were
taken forward to cost-sensitive learning optimisation.One
hundred and ninety-two patients experienced acute
desquamation. Resampling and cost-sensitive learning
optimisation facilitated an improvement in classification
performance. Based on maximising sensitivity (true
positives), the 'hero' model was the cost-sensitive random
forest algorithm with a false-negative: false-positive
misclassification penalty of 90:1 containing m = 114
predictive features. Model sensitivity and specificity were
0.77 and 0.66, respectively, with an area under the curve of
0.77 in the validation cohort.ML algorithms with resampling
and cost-sensitive learning generated clinically valid
prediction models for acute desquamation using patient
demographic and treatment features. Further external
validation and inclusion of genomic markers in ML prediction
models are worthwhile, to identify patients at increased
risk of toxicity who may benefit from supportive
intervention or even a change in treatment plan.},
cin = {C020},
ddc = {610},
cid = {I:(DE-He78)C020-20160331},
pnm = {313 - Krebsrisikofaktoren und Prävention (POF4-313)},
pid = {G:(DE-HGF)POF4-313},
typ = {PUB:(DE-HGF)16},
pubmed = {pmid:35647396},
pmc = {pmc:PMC9133391},
doi = {10.1016/j.adro.2021.100890},
url = {https://inrepo02.dkfz.de/record/180190},
}