Metric Learning on Healthcare Data with Incomplete Modalities

Authors: Qiuling Suo, Weida Zhong, Fenglong Ma, Ye Yuan, Jing Gao, Aidong Zhang

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that the proposed framework learns more accurate distance metric on real-world healthcare datasets with incomplete modalities, comparing with the state-of-the-art approaches.
Researcher Affiliation Academia 1Department of Computer Science and Engineering, SUNY at Buffalo, NY, USA 2College of Information and Communication Engineering, Beijing University of Technology, China 3Department of Computer Science, University of Virginia, VA, USA
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes The Alzheimer s Disease Neuroimaging Initiative (ADNI) 1 project aims to track the progression of Alzheimer s disease using biomarkers and clinical measures. There are 840 patients in three cohorts: 231 cognitively normal (CN), 410 mild cognitive impairment (MCI), and 199 Alzheimer s disease (AD) patients. In this work, we use the available modalities in the database: MRI images and PET images, and generate the missing images from each other. and 1https://adni.loni.usc.edu/
Dataset Splits Yes We randomly divide the patient set into training, validation and testing sets in a 0.75:0.05:0.2 ratio.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments.
Software Dependencies No The paper mentions 'SPM2 and CAT12' softwares but does not provide specific version numbers for these or other key software dependencies used in their implementation.
Experiment Setup No The paper states "We set the learning rate and the network structures the same as [Cai et al., 2018] but in a 2D fashion," but it does not explicitly list these concrete hyperparameter values or network structure details within its own main text.