DPSCREEN: Dynamic Personalized Screening
Authors: Kartik Ahuja, William Zame, Mihaela van der Schaar
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | As an illustration, we make use of a large breast cancer data set. The constructed policy screens patients more or less often according to their initial risk it is personalized to the features of the patient and according to the results of previous screens it is personalized to the history of the patient. In comparison with existing clinical policies, the constructed policy leads to large reductions (28-68%) in the number of screens performed while achieving the same expected delays in disease detection. |
| Researcher Affiliation | Academia | Kartik Ahuja Electrical and Computer Engineering Department University of California, Los Angeles ahujak@ucla.edu William R. Zame Economics Department University of California, Los Angeles zame@econ.ucla.edu Mihaela van der Schaar Engineering Science Department, University of Oxford Electrical and Computer Engineering Department, University of California, Los Angeles mihaela.vanderschaar@oxford-man.ox.ac.uk |
| Pseudocode | Yes | In Algorithm 1 (pseudo-code in the Appendix A of the Supplementary Materials), we first construct a lower dimensional belief space by sampling trajectories that are more likely to occur for the disease and then sampling the set of beliefs in the lower dimensional space that are likely to occur over the course of various screening policies. The key steps for Algorithm 1 are [...] Denote the set of belief vectors constructed at time t by B[t] and the set of all such beliefs as B = { B[t], t}. We carry out point-based value backups on these beliefs B (see Algorithm 2 in the Appendix A of the Supplementary Materials), to construct the alpha vectors and thus the approximately optimal policy. |
| Open Source Code | No | The paper does not provide any concrete access information (link or explicit statement) to open-source code for the described methodology. |
| Open Datasets | Yes | We use a de-identified dataset (from Athena Health Network [22]) of 45, 000 patients aged 60-65 who underwent screening for breast cancer. For most individuals we have the following associated features: age, the number of family members with breast cancer, weight, etc. Each patient had at least one mammogram; some had several. (In total, there are 84,000 mammograms in the dataset.) |
| Dataset Splits | No | The paper describes using a dataset and dividing it into risk groups but does not specify explicit training, validation, or test dataset splits (e.g., percentages, sample counts, or predefined citations). |
| Hardware Specification | No | The paper does not provide any specific hardware details (e.g., GPU/CPU models, processor types, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions several models and methods (e.g., Cox Proportional Odds model, Gail model, Markov Chain Monte Carlo methods) but does not provide specific software dependencies with version numbers (e.g., library names with versions) needed to replicate the experiment. |
| Experiment Setup | Yes | We assumed pin(x) and ptran(x) are logistic functions of x. We use standard Markov Chain Monte Carlo methods to estimate these functions pin(x) and ptran(x) (further details in the Appendix G of the Supplementary Materials). We assume that each woman has one self-examination per month [25] [26]. We use the value ι = 0.9 as stated in [23]. We estimate the parameters for the self-examinations σ = 0.43 and y = 1 on the basis of the values of sensitivity and specificity for the self-examination from the literature [43]. |