Learning Adaptive Forecasting Models from Irregularly Sampled Multivariate Clinical Data
Authors: Zitao Liu, Milos Hauskrecht
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The proposed forecasting model is evaluated on a real-world clinical time series dataset. The results demonstrate that our approach is superior on the prediction tasks for multivariate, irregularly sampled clinical time series, and it outperforms both the population based and patient-speciļ¬c time series prediction models in terms of prediction accuracy. |
| Researcher Affiliation | Academia | Zitao Liu and Milos Hauskrecht Computer Science Department University of Pittsburgh 210 South Bouquet St., Pittsburgh, PA, 15260 USA |
| Pseudocode | Yes | Algorithm 1 summarizes our two-stage adaptive forecasting model and its learning and prediction parts. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. It mentions supplementary material at a URL, but the linked file is a PDF, not source code. |
| Open Datasets | Yes | We test our two-stage adaptive model on a clinical MTS data obtained from EHRs of post-surgical cardiac patients in PCP database (Hauskrecht et al. 2010; 2013). |
| Dataset Splits | No | The paper mentions 'randomly selected 100 patients out of 500 as a test set and used the remaining 400 patients for training the models' but does not specify a separate validation split or explicit cross-validation for evaluation. It mentions 'internal cross validation approach' for hyperparameter selection, but not for a dataset split for validation. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions various models and algorithms (LDS, GP, MTGP, EM algorithm, Kalman filtering) but does not provide specific version numbers for any software, libraries, or frameworks used in the implementation. |
| Experiment Setup | Yes | We would also like to note that the hyper parameters (e.g., DVI sampling frequency r, number of hidden states in LDS d) used in our methods are selected (in all experiments) by the internal cross validation approach while optimizing models predictive performance. Given the short span characteristic and the fact that the median length of our clinical data is 14, we predict the last four observations for each patient dynamic in test data. |