EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Can Data Augmentation Change the Best-Performing Classifier? Evidence From a Machine Learning Study on Sarcopenia

Kann Data Augmentation den bestperformenden Klassifikator verändern? Evidenz aus einer Machine-Learning-Studie zu Sarkopenie

MARIA DE COLA, FABIANA PIA VITIELLO, Chiara Scarfì, Elisabetta Morini, Mattia Papa, Danilo Caudo, Davide Sattin, Antonio Musaro, Nunzio Iraci, Christian Lunetta

Epidemiology, Biostatistics, and Public Health · 22.09.2026

DOI: 10.54103/2282-0930/32091 ↗

Auf einen Blick

In einer Machine-Learning-Studie zu Sarkopenie war KNN auf den Originaldaten am besten. Nach Data Augmentation änderte sich der bestperformende Klassifikator je nach Setting.

Fragestellung

Bleibt der auf Originaldaten beste Klassifikator nach Data Augmentation der leistungsstärkste?

Methode

Vergleich von 5 überwachten Klassifikationsmodellen auf BIA- und Ultraschall-Daten; 50 Personen; Validierung per Leave-One-Out-Cross-Validation; Augmentation durch jittering.

Ergebnisse

Auf Originaldaten war KNN am besten: balanced accuracy 0,650, recall 0,600, precision 0,571, F1-Score 0,585. Nach Augmentation erreichte SVM die beste Leistung: balanced accuracy 0,795, recall 0,741, precision 0,769, F1-Score 0,755. Bei niedrigerem Threshold war LR-L2 + KNN am besten (balanced accuracy 0,807, recall 0,815, precision 0,733, F1-Score 0,772).

Schlussfolgerung der Autor:innen

Die Ergebnisse sprechen dafür, dass der beste Klassifikator auf Originaldaten nach Data Augmentation nicht unbedingt der beste bleibt. Die Autor:innen betonen die erneute Bewertung der Modelle nach Datenaugmentation.

Kernaussagen

  • KNN war auf Originaldaten am besten.
  • Nach Augmentation lag SVM vorn.
  • Bei niedrigerem Threshold wurde wieder KNN bzw. LR-L2 + KNN am besten.

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Epidemiology, Biostatistics, and Public Health, online erschienen am 22.09.2026. Verfasst von MARIA DE COLA und 9 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

Auf einen BlickAbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

One of the M-Brain project's aims is to identify biomarkers that can distinguish between healthy and pathological ageing. The choice of classifier can have a significant impact on predictive performance due to the challenges posed by limited sample size, class imbalance and heterogeneous features. Using data augmentation techniques can improve the performance of different machine learning algorithms.

Objectives

Compare different supervised classification models to assess whether the best-performing classifier on real data remained the most effective after data augmentation.

Methods

We used bioelectrical impedance analysis (BIA) parameters and muscle thickness data measured by ultrasound collected. The original dataset included 50 subjects (20 diagnosed with sarcopenia according to EWGSOP2 criteria, and 30 without) and 17 variables (nine for BIA and eight for ultrasound). We trained five supervised classification algorithms—Logistic Regression with L2 regularization (LR-L2), K-Nearest Neighbors (KNN), Support Vector Machine (SVM) with a radial basis function kernel, Linear Discriminant Analysis, and Random Forest—on the original dataset. The dataset was subsequently augmented using jittering, a technique that generates synthetic samples through small random perturbations of the training data. To preserve the underlying data structure, synthetic instances were generated using the learned weights of each classifier. Any new instance was added to the dataset according to a given threshold. The five algorithms were then re-evaluated to assess the impact of data augmentation on classification performance. Model validation was performed using Leave-One-Out Cross-Validation, and performance by balanced accuracy, recall, precision, and F1-score measures.

Results

In the real-data analysis, KNN showed the best overall performance, achieving a balanced accuracy of 0.650, recall of 0.600, precision of 0.571, and F1-score of 0.585. All classifiers performance improved after data augmentation, although gains were not consistent across all models and settings. The KNN + SVM pipeline performed best, i.e. the feature weights derived from KNN guided the selection of the most suitable synthetic instances generated through jittering, when we used a conservative threshold (1.0–0.7). The five classifiers were then trained and evaluated on the augmented dataset, and SVM achieved the best performance, reaching a balanced accuracy of 0.795, recall of 0.741, precision of 0.769, and F1-score of 0.755. Surprisingly, the LR-L2 + KNN pipeline achieved the highest classification metrics at the lower threshold (0.6–0.5), reaching a balanced accuracy of 0.807, recall of 0.815, precision of 0.733, and F1-score of 0.772 at a threshold of 0.5.

Conclusions

The results suggest that the classifier that performs best on the original dataset may not remain the best after data augmentation. KNN achieved the highest performance on the original data, whereas SVM outperformed all other classifiers after synthetic data generation. However, when the threshold for generating synthetic instances was relaxed, KNN once again became the top-performing classifier. These findings underscore the importance of reevaluating classification models after data augmentation rather than assuming that the model selected on the original dataset will continue to perform best. They also highlight the importance of balancing class distributions and verifying that the statistical properties of the original variables are preserved following data augmentation.

Zitieren

APA

COLA, M. D., VITIELLO, F. P., Scarfì, C., Morini, E., Papa, M., Caudo, D., Sattin, D., Musaro, A., Iraci, N., & Lunetta, C. (2026). Can Data Augmentation Change the Best-Performing Classifier? Evidence From a Machine Learning Study on Sarcopenia. Epidemiology, Biostatistics, and Public Health. https://doi.org/10.54103/2282-0930/32091

Vancouver

COLA MD, VITIELLO FP, Scarfì C, Morini E, Papa M, Caudo D, et al. Can Data Augmentation Change the Best-Performing Classifier? Evidence From a Machine Learning Study on Sarcopenia. Epidemiology, Biostatistics, and Public Health. 2026. doi:10.54103/2282-0930/32091

BibTeX
@article{COLA2026,
  author = {COLA, MARIA DE and VITIELLO, FABIANA PIA and Scarfì, Chiara and Morini, Elisabetta and Papa, Mattia and Caudo, Danilo and Sattin, Davide and Musaro, Antonio and Iraci, Nunzio and Lunetta, Christian},
  title = {Can Data Augmentation Change the Best-Performing Classifier? Evidence From a Machine Learning Study on Sarcopenia},
  journal = {Epidemiology, Biostatistics, and Public Health},
  year = {2026},
  doi = {10.54103/2282-0930/32091}
}
Kostenlos starten →