Transformer-Based Video-Structure Multi-Instance Learning for Whole Slide Image Classification

Authors: Yingfan Ma, Xiaoyuan Luo, Kexue Fu, Manning Wang

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on three public WSI datasets show that our proposed method outperforms existing SOTA methods in both WSI classification and positive region detection.
Researcher Affiliation Academia Yingfan Ma1 2, Xiaoyuan Luo1 2, Kexue Fu3, Manning Wang1 2 * 1Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, Shanghai 200032, China 2Shanghai Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention, Shanghai 200032, China 3Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Shandong Computer Science Center (National Supercomputer Center in Jinan), Jinan, China 22211010089@m.fudan.edu.cn, {19111010030, mnwang}@fudan.edu.cn, fukx@sdas.org
Pseudocode No The paper describes its methods in prose and through figures, but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not include any explicit statements about releasing its source code nor does it provide a link to a code repository for the methodology described.
Open Datasets Yes Extensive experiments on three public WSI datasets: CAMELYON16, PANDA, and TCGA-NSCLC. ... CAMELYON16 (Bejnordi et al. 2017), PANDA (Bulten et al. 2022). ... TCGA-NSCLC1 ... 1http://www.cancer.gov/tcga
Dataset Splits No The paper mentions using 'training set' and 'test set' and conducting 'ablation studies' but does not provide specific percentages, sample counts, or explicit methodology for how the data was split into training, validation, and test sets for the main experiments.
Hardware Specification Yes All experiments are conducted using 2 A100s.
Software Dependencies No The paper mentions using an 'Adam optimizer' and 'Res Net18' as an encoder but does not provide specific version numbers for programming languages or software libraries like PyTorch or TensorFlow.
Experiment Setup Yes During the training process, we utilized the cross-entropy loss, with an Adam optimizer having a learning rate of 1e-4 and weight decay of 1e-4.