Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Lymphoma Sub-type Classification Using DenseNet169-based Transfer Learning Architecture

Rektorat UDINUS 2024 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2024
Goals
Record
17cf48a6-78b9-4966-969a-5032d76705de
Published
2026-10-30

Universitas Dian Nuswantoro conducted a study to classify lymphoma cells using a digital pathology dataset. The dataset comprised three types: Chronic Lymphocytic Leukemia (CLL), Follicular Lymphoma (FL), and Mantle Cell Lymphoma (MCL), with 113, 139, and 122 images, respectively. The activity involved UDINUS researchers Riyanto, Azizu Ahmad Rozaki, Ardytha Luthfiarta, Farrel Ardra Muhammad, Yohanes Deny Novandian, Muhammad Hafizh Dzaki, Ika Novita Dewi, Yani Parti Astuti, Rismiyati, and Ni Kadek Devi Adnyaswari Putri.

The study addressed lymphoma, which is described as a serious disease with significant potential complications, including heart disease, lung disease, and increased susceptibility to infections due to immune system suppression. To address the limited dataset, the activity applied image augmentation techniques. The images were resized to 40% of their original dimensions of 1388 x 1040 pixels. Classification used a deep learning approach with transfer learning, specifically the DenseNet169 model from TensorFlow. The model was initialized with ImageNet weights and optimized with specific parameters.

The DenseNet169 model achieved an accuracy of 99.1%, surpassing previous research by 0.4%. The study reports that image augmentation combined with transfer learning effectively enhances classification accuracy in the context of limited datasets. This activity is framed as UDINUS health research using digital pathology and deep learning to classify the three lymphoma types represented in the dataset.

Reference https://www.scopus.com/pages/publications/85213311266?origin=resultslist