SavedAdd your journal

Journals / Türkiye Teknoloji ve Uygulamalı Bilimler Dergisi / 2026 / Sayı: 2026

Multidimensional Performance and Model Selection Analysis of Transfer Learning-Based Deep Learning Models on Multiple Medical Imaging Datasets

Transfer Öğrenme Tabanlı Derin Öğrenme Modellerinin Çoklu Tıbbi Görüntü Veri Setlerinde Çok Boyutlu Performans ve Model Seçim Analizi

Lokman Doğan(KÜTAHYA SAĞLIK BİLİMLERİ ÜNİVERSİTESİ, MESLEK YÜKSEKOKULU)0000-0002-5039-3418
Pages
156–166
Publication Date
2026-10-01
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The experimental results showed that model performance varied depending on the dataset; all models achieved very high classification accuracy on the BloodMNIST dataset, while the DermaMNIST dataset proved to be the most challenging for the models.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Transfer learning-based deep learning models have been widely used in recent years because they can achieve high classification accuracy with a limited amount of labeled medical image data.
📊
Key Findings & Data:However, there are few comparative studies that jointly evaluate the accuracy, computational efficiency, and performance consistency across different medical image datasets of current transfer learning models.
💡
Conclusion & Impact:The experimental results showed that model performance varied depending on the dataset; all models achieved very high classification accuracy on the BloodMNIST dataset, while the DermaMNIST dataset proved to be the most challenging for the models.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Transfer öğrenme · derin öğrenme · tıbbi görüntü sınıflandırma

Abstract

Transfer learning-based deep learning models have been widely used in recent years because they can achieve high classification accuracy with a limited amount of labeled medical image data. However, there are few comparative studies that jointly evaluate the accuracy, computational efficiency, and performance consistency across different medical image datasets of current transfer learning models. In this study, the performance of the ResNet50, DenseNet121, EfficientNetB0, EfficientNetV2B0, MobileNetV3, and ConvNeXt-Tiny models, using a transfer learning approach, was comprehensively compared across the PneumoniaMNIST, DermaMNIST, and BloodMNIST datasets. All models were trained on Google Colab using an NVIDIA A100-SXM4-80GB GPU, with the same training parameters, the same data splits, and the same experimental environment. Training utilized the AdamW optimization algorithm, a learning rate of 1×10⁻⁴, a batch size of 32, a maximum of 10 epochs, an early stopping mechanism, and a fixed random seed of 42. The models were evaluated in terms of accuracy, precision, recall, macro F1-score, ROC-AUC, training time, inference time, frames per second (FPS), model size, and number of parameters. The experimental results showed that model performance varied depending on the dataset; all models achieved very high classification accuracy on the BloodMNIST dataset, while the DermaMNIST dataset proved to be the most challenging for the models. In the overall evaluation, the DenseNet121 and ConvNeXt-Tiny models demonstrated the highest average classification performance, while the MobileNetV3 and EfficientNetB0 models stood out in terms of computational efficiency with their smaller model sizes and faster inference times. The findings provide guidance to practitioners and researchers on selecting the appropriate transfer learning model for various medical image classification problems.

Özet

Transfer öğrenme tabanlı derin öğrenme modelleri, sınırlı miktarda etiketli tıbbi görüntü verisi ile yüksek sınıflandırma başarısı sağlayabilmeleri nedeniyle son yıllarda yaygın olarak kullanılmaktadır. Bununla birlikte farklı tıbbi görüntü veri setlerinde güncel transfer öğrenme modellerinin doğruluk, hesaplama verimliliği ve veri setleri arasındaki performans tutarlılığını birlikte değerlendiren karşılaştırmalı çalışmalar sınırlıdır. Bu çalışmada ResNet50, DenseNet121, EfficientNetB0, EfficientNetV2B0, MobileNetV3 ve ConvNeXt Tiny modellerinin transfer öğrenme yaklaşımı kullanılarak PneumoniaMNIST, DermaMNIST ve BloodMNIST veri setleri üzerindeki performansları kapsamlı olarak karşılaştırılmıştır. Tüm modeller aynı eğitim parametreleri, aynı veri bölünmeleri ve aynı deney ortamında Google Colab üzerinde NVIDIA A100-SXM4-80GB GPU kullanılarak eğitilmiştir. Eğitimlerde AdamW optimizasyon algoritması, 1×10⁻⁴ öğrenme oranı, 32 batch büyüklüğü, maksimum 10 epoch, erken durdurma mekanizması ve sabit 42 rastgelelik tohumu kullanılmıştır. Modeller; Accuracy, Precision, Recall, Macro F1-score, ROC-AUC, eğitim süresi, çıkarım süresi, saniyedeki kare sayısı (FPS), model boyutu ve parametre sayısı açısından değerlendirilmiştir. Deneysel sonuçlar, veri setine bağlı olarak model performanslarının değiştiğini, BloodMNIST veri setinde tüm modellerin oldukça yüksek sınıflandırma başarısı elde ettiğini, DermaMNIST veri setinin ise modeller için en zorlayıcı veri seti olduğunu göstermiştir. Genel değerlendirmede DenseNet121 ve ConvNeXt Tiny modelleri en yüksek ortalama sınıflandırma performansını sergilerken, MobileNetV3 ve EfficientNetB0 modelleri daha düşük model boyutu ve daha hızlı çıkarım süreleriyle hesaplama verimliliği açısından öne çıkmıştır. Elde edilen bulgular, farklı tıbbi görüntü sınıflandırma problemleri için uygun transfer öğrenme modelinin seçimine yönelik uygulayıcılara ve araştırmacılara yol gösterici niteliktedir.

🔖

Cite This Article

Doğan, L. (2026). Multidimensional Performance and Model Selection Analysis of Transfer Learning-Based Deep Learning Models on Multiple Medical Imaging Datasets. Türkiye Teknoloji ve Uygulamalı Bilimler Dergisi, Sayı: 2026, 156–166. https://doi.org/10.70562/tubid.2000572

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (17)

  1. Doğan, Y. and Özdemir, C. (2024). Enhancing skin cancer diagnosis through the integration of deep learning and machine learning approaches. Bilişim Teknolojileri Dergisi, 17(4), 339-347.
  2. Doğan, Y., Özdemir, C. and Kaya, Y. (2024). Enhancing CNN model classification performance through RGB angle rotation method. Neural Computing and Applications, 36(32), 20259-20276.
  3. Esteva, A., Chou, K., Yeung, S., Naik, N., Madani, A., Mottaghi, A., Liu, Y., Topol, E., Dean, J. and Socher, R. (2021). Deep learning-enabled medical computer vision. npj Digital Medicine, 4(1), 5.
  4. He, K., Zhang, X., Ren, S. and Sun, J. (2016). 'Deep residual learning for image recognition', 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA, 27-30 June 2016. IEEE, 770-778.
  5. Howard, A., Sandler, M., Chen, B., Wang, W., Chen, L., Tan, M., Chu, G., Vasudevan, V., Zhu, Y., Pang, R., Adam, H. and Le, Q. (2019). 'Searching for MobileNetV3', 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, South Korea, 27 October-2 November 2019. IEEE, 1314-1324.
  6. Huang, G., Liu, Z., Van Der Maaten, L. and Weinberger, K. Q. (2017). 'Densely connected convolutional networks', 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, 21-26 July 2017. IEEE, 2261-2269.
  7. Kather, J. N., Pearson, A. T., Halama, N., Jäger, D., Krause, J., Loosen, S. H., Marx, A., Boor, P., Tacke, F., Neumann, U. P., Grabsch, H. I., Yoshikawa, T., Brenner, H., Chang-Claude, J., Hoffmeister, M., Trautwein, C. and Luedde, T. (2019). Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer. Nature Medicine, 25(7), 1054-1056.
  8. Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., van der Laak, J. A. W. M., van Ginneken, B. and Sánchez, C. I. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42(1), 60-88.
  9. Liu, Z., Mao, H., Wu, C., Feichtenhofer, C., Darrell, T. and Xie, S. (2022). 'A ConvNet for the 2020s', 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, USA, 19-24 June 2022. IEEE, 11966-11976.
  10. Oral, S., Ökten, İ. and Yüzgeç, U. (2023). Fungus classification based on CNN deep learning model. Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, 12(1), 226-241.
  11. Özdemir, C. (2023). Designing effective models for COVID-19 diagnosis through transfer learning and interlayer visualization. Balkan Journal of Electrical and Computer Engineering, 11(4), 340-345.
  12. Pan, S. J. and Yang, Q. (2010). A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345-1359.
  13. Talo, M. (2019). Automated classification of histopathology images using transfer learning. Artificial Intelligence in Medicine, 101(1), 101743.
  14. Tan, M. and Le, Q. V. (2019). 'EfficientNet: Rethinking model scaling for convolutional neural networks', 36th International Conference on Machine Learning (ICML), Long Beach, USA, 9-15 June 2019. PMLR, 6105-6114.
  15. Tan, M. and Le, Q. V. (2021). 'EfficientNetV2: Smaller models and faster training', 38th International Conference on Machine Learning (ICML), Virtual, 18-24 July 2021. PMLR, 10096-10106.
  16. Wightman, R. (2025). PyTorch Image Models (timm). [online] Available at: https://github.com/huggingface/pytorch-image-models [Accessed: 15 June 2026].
  17. Yang, J., Shi, R., Wei, D., Liu, Z., Zhao, L., Ke, B., Pfister, H. and Ni, B. (2023). MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification. Scientific Data, 10(1), 41.