Back to Search Start Over

Automated data preparation for in vivo tumor characterization with machine learning.

Automated data preparation for in vivo tumor characterization with machine learning.

Authors :
Krajnc D
Spielvogel CP
Grahovac M
Ecsedi B
Rasul S
Poetsch N
Traub-Weidinger T
Haug AR
Ritter Z
Alizadeh H
Hacker M
Beyer T
Papp L
Source :
Frontiers in oncology [Front Oncol] 2022 Oct 11; Vol. 12, pp. 1017911. Date of Electronic Publication: 2022 Oct 11 (Print Publication: 2022).
Publication Year :
2022

Abstract

Background: This study proposes machine learning-driven data preparation (MLDP) for optimal data preparation (DP) prior to building prediction models for cancer cohorts.<br />Methods: A collection of well-established DP methods were incorporated for building the DP pipelines for various clinical cohorts prior to machine learning. Evolutionary algorithm principles combined with hyperparameter optimization were employed to iteratively select the best fitting subset of data preparation algorithms for the given dataset. The proposed method was validated for glioma and prostate single center cohorts by 100-fold Monte Carlo (MC) cross-validation scheme with 80-20% training-validation split ratio. In addition, a dual-center diffuse large B-cell lymphoma (DLBCL) cohort was utilized with Center 1 as training and Center 2 as independent validation datasets to predict cohort-specific clinical endpoints. Five machine learning (ML) classifiers were employed for building prediction models across all analyzed cohorts. Predictive performance was estimated by confusion matrix analytics over the validation sets of each cohort. The performance of each model with and without MLDP, as well as with manually-defined DP were compared in each of the four cohorts.<br />Results: Sixteen of twenty established predictive models demonstrated area under the receiver operator characteristics curve (AUC) performance increase utilizing the MLDP. The MLDP resulted in the highest performance increase for random forest (RF) (+0.16 AUC) and support vector machine (SVM) (+0.13 AUC) model schemes for predicting 36-months survival in the glioma cohort. Single center cohorts resulted in complex (6-7 DP steps) DP pipelines, with a high occurrence of outlier detection, feature selection and synthetic majority oversampling technique (SMOTE). In contrast, the optimal DP pipeline for the dual-center DLBCL cohort only included outlier detection and SMOTE DP steps.<br />Conclusions: This study demonstrates that data preparation prior to ML prediction model building in cancer cohorts shall be ML-driven itself, yielding optimal prediction models in both single and multi-centric settings.<br />Competing Interests: MH, LP, and TB are co-founders of Dedicaid GmbH, Austria. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.<br /> (Copyright © 2022 Krajnc, Spielvogel, Grahovac, Ecsedi, Rasul, Poetsch, Traub-Weidinger, Haug, Ritter, Alizadeh, Hacker, Beyer and Papp.)

Details

Language :
English
ISSN :
2234-943X
Volume :
12
Database :
MEDLINE
Journal :
Frontiers in oncology
Publication Type :
Academic Journal
Accession number :
36303841
Full Text :
https://doi.org/10.3389/fonc.2022.1017911