issues Search Results · language:Dune language:JavaScript language:JavaScript language:Python language:JavaScript linked:pr
Filter by
4.5M results
Support ITT vs per-protocol estimands and prespecified subgroup / effect-modification analyses per trial. Acceptance:
estimand is selectable; subgroup results reported.
phase-2
Jpickard1
Add E-values + negative-control outcomes/exposures + quantitative bias analysis to every emulation, for
unmeasured-confounding robustness. Acceptance: each emulation reports E-value + at least one negative-control ...
phase-2
Run the full pipeline over all emulable trials x datasets and assemble a results gallery (emulated vs observed for
each). The flagship breadth deliverable. Acceptance: a generated gallery covering the ...
phase-2
Scan the ctgov corpus, score which trials are emulable in each dataset (data availability for
eligibility/exposure/outcome), and produce a ranked emulable-trial catalog. Acceptance: a sortable catalog ...
phase-2
Quantify missing data + the proxy variables used per trial-per-DB and their impact on the estimate. Acceptance:
missingness summary + proxy list + sensitivity to proxies.
phase-2
For each trial, classify every protocol element (eligibility, exposure, outcomes, covariates) as measurable / proxy /
unmeasurable in each dataset, and surface the gaps. Explains what we can definitively ...
phase-2
For each trial, compare the emulated effect across eICU/MIMIC/MGB with heterogeneity stats (I-squared) + forest plot,
and attribute differences (cohort, coding, missingness). Acceptance: per-trial forest ...
phase-2
Harmonize ICU vs hospital timescales, lookback/washout windows, and per-DB timestamp precision into one timing contract
so estimands are comparable across eICU/MIMIC/MGB. Acceptance: a timing-config drives ...
phase-2
Formal time-zero alignment with grace-window landmark + clone-censor-weight option to remove immortal-time bias.
Acceptance: time-zero is explicit per strategy; immortal-time sensitivity reported.
phase-2
Per trial-per-dataset, emit a CONSORT-style flow (screened to eligible to enrolled to analyzed) + cohort diagnostics
(baseline table, attrition reasons). Makes cohort definitions transparent + reproducible. ...
phase-2
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.