AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning
Authors: Ahmed Alaa, Mihaela Schaar
ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conclude the paper by conducting a set of experiments on multiple patient cohorts representing various aspects of cardiovascular patient care, and show that prognostic models learned by AUTOPROGNOSIS outperform widely used clinical risk scores and existing Auto ML frameworks. |
| Researcher Affiliation | Academia | 1University of California, Los Angeles, USA 2University of Oxford, Oxford, UK 3Alan Turing Institute, London, UK. |
| Pseudocode | No | The paper includes a schematic depiction of AUTOPROGNOSIS (Figure 2) and describes algorithmic steps in prose, but it does not provide any formal pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain any explicit statements or links indicating that the source code for AUTOPROGNOSIS is publicly available. |
| Open Datasets | Yes | We considered a major cohort for preventive cardiology: the Meta-analysis Global Group in Chronic heart failure database (MAGGIC), which holds data for 46,817 patients gathered from multiple clinical studies (Wong et al., 2014). |
| Dataset Splits | Yes | The main goal of AUTOPROGNOSIS is to identify the best pipeline configuration P θ PΘ for a given patient cohort D via J-fold cross-validation as follows: P θ arg max Pθ PΘ 1 J J i=1 L(Pθ; D(i) train, D(i) valid) |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper mentions software like scikit-learn and XGBoost as part of the pipeline but does not provide specific version numbers for any software dependencies. |
| Experiment Setup | Yes | All algorithms were allowed to run for a maximum of 10 hours to ensure a fair comparison. |