Medical Concept Embedding with Multiple Ontological Representations
Authors: Lihong Song, Chin Wang Cheong, Kejing Yin, William K. Cheung, Benjamin C. M. Fung, Jonathan Poon
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We apply MMORE to diagnosis prediction and our experimental results show that the representations obtained by MMORE can achieve better predictive accuracy and result in clinically meaningful sub-categorizations of the existing ontological categories. We conduct experiments based on the MIMIC-III dataset to compare the performance of our proposed method MMORE with several state-of-the-art methods in terms of the prediction accuracy for the next-admission diagnosis prediction. |
| Researcher Affiliation | Collaboration | Lihong Song1 , Chin Wang Cheong1 , Kejing Yin1 , William K. Cheung 1 , Benjamin C. M. Fung2 and Jonathan Poon3 1Department of Computer Science, Hong Kong Baptist University, Hong Kong SAR, China 2School of Information Studies, Mc Gill University, Montreal, Canada 3Hong Kong Hospital Authority, Hong Kong SAR, China |
| Pseudocode | No | The paper describes its model and methods through mathematical equations and textual explanations, but it does not include any formal 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code for the methodology, nor does it include links to a code repository. |
| Open Datasets | Yes | We evaluated our proposed model using the open-source MIMIC-III dataset [Johnson et al., 2016]. Data set. MIMIC-III (Medical Information Mart for Intensive Care) [Johnson et al., 2016] is a open-source dataset which comprises over 46k de-identified ICU patients collected over 11 years. |
| Dataset Splits | Yes | We randomly split the data into training set, validation set and test set, and fix the size of the validation set to be 10%. To validate the robustness against insufficient data, we vary the size of the training set from 20% to 80% and use the remaining part as the test set. |
| Hardware Specification | No | The paper does not specify any details about the hardware used for running the experiments (e.g., CPU, GPU models, memory). |
| Software Dependencies | No | The paper mentions using Adadelta for optimization but does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions or other libraries). |
| Experiment Setup | Yes | We set the dimension of both the ontological embedding and the co-occurrence embedding to be 400 in our model. The embedding dimension of all baselines are set to be 800 for fair comparison as our model concatenates the two embedding matrices. The dimension of the hidden layer in the perceptrons used for the attention mechanism are set to be 100. The model is optimized using Adadelta [Zeiler, 2012] with batch size of 100. |