Building Better Deep Learning Models Through Dataset Fusion: A Case Study in Skin Cancer Classification with Hyperdatasets
Diagnostics · 03.02.2025 · 15 (3) · 352
DOI: 10.3390/diagnostics15030352 ↗Kurz gefasst
Fachartikel in der Zeitschrift Diagnostics, online erschienen am 03.02.2025. Verfasst von Panagiotis Georgiadis und 4 weiteren Autor:innen. Beteiligte Einrichtung: MLV Research Group, Department of Informatics, Democritus University of Thrace, 65404 Kavala, Greece. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background/Objectives: This work brings to light the importance of forming large training datasets with diverse images generated and proposes an image dataset merging application, namely, the Data Merger App, to streamline the management and synthesis of large-scale datasets. The Data Merger can recognize common classes across various datasets and provides tools to combine and organize them in a well-structured and easily accessible way.
Methods
A case study is then presented, leveraging four different Convolutional Neural Network (CNN) models, VGG16, ResNet50, MobileNetV3-small, and DenseNet-161, and a Visual Transformer (ViT), to benchmark their performance to classify skin cancer images, when trained on single datasets and on enhanced hyperdatasets generated by the Data Merger App.
Results
Extended experimental results indicated that enhanced hyperdatasets are efficient and able to improve the accuracies of classification models, whether the models are trained from scratch or by using Transfer Learning. Moreover, the ViT model was reported for higher classification accuracies compared to CNNs on datasets with a limited number of classes, reporting 91.87% accuracy for 9 classes, as well as in the case of enhanced hyperdatasets with multiple numbers of classes, reporting accuracy of 58% for 32 classes.
Conclusions
In essence, this work demonstrates the great significance of data combination, as well as the utility value of the developed prototype web application as a critical tool for researchers and data scientists, enabling them to easily handle complex datasets, combine datasets into larger diverse versions, to further enhance the generalization ability of models and improve the quality and impact of their work.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- MLV Research Group, Department of Informatics, Democritus University of Thrace, 65404 Kavala, Greece.
ORCID-Profile: Eleni Vrochidou, George A Papakostas
Zitieren
APA
Georgiadis, P., Gkouvrikos, E. V., Vrochidou, E., Kalampokas, T., & Papakostas, G. A. (2025). Building Better Deep Learning Models Through Dataset Fusion: A Case Study in Skin Cancer Classification with Hyperdatasets. Diagnostics, 15(3), 352. https://doi.org/10.3390/diagnostics15030352
Vancouver
Georgiadis P, Gkouvrikos EV, Vrochidou E, Kalampokas T, Papakostas GA. Building Better Deep Learning Models Through Dataset Fusion: A Case Study in Skin Cancer Classification with Hyperdatasets. Diagnostics. 2025;15(3):352. doi:10.3390/diagnostics15030352
BibTeX
@article{Georgiadis2025,
author = {Georgiadis, Panagiotis and Gkouvrikos, Emmanouil V and Vrochidou, Eleni and Kalampokas, Theofanis and Papakostas, George A},
title = {Building Better Deep Learning Models Through Dataset Fusion: A Case Study in Skin Cancer Classification with Hyperdatasets},
journal = {Diagnostics},
year = {2025},
volume = {15},
number = {3},
pages = {352},
doi = {10.3390/diagnostics15030352}
}
EUVIMED