Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -124,14 +124,14 @@ EHR_TYPES:
- $t_s$ : time point marking the date of the scan.
- $t_a$ : time point marking the ```DAYS AFTER``` a scan (e.g., window)
- $t_h$ : time point marking the last date of record in a patient's history.
> Note that $t_a$ - $t_s$, which is the window in which we're identifying events specified for our experimental design, may be longer than $t_h$ - $t_0$, which is the time period a patient has data in the EHR for. This is crucial!
> Note that $t_a$ - $t_s$, which is the window in which we're identifying events specified for our experimental design, maybe longer than $t_h$ - $t_0$, which is the time period a patient has data in the EHR for. This is crucial!
#### Output Label Interpretation
- **Class 0** → Patient not diagnosed with the specified disease before or after the CT Scan within the requested time window or any time in the patient's history.
- IF a patient develops disease, then $t_d > t_a$.
- IF a patient develops a disease, then $t_d > t_a$.
- **Class 1** → Patient diagnosed with specified disease during the time window between days before and days after.
- Commontly, only patients from Class 0 and 1 are included in the studies as controls/positive classes.
- Commonly, only patients from Class 0 and 1 are included in the studies as controls/positive classes.
- $t_b ≤ t_d ≤ t_a$
- **Class 2** → Patient diagnosed with specified disease earlier than the 'days before' window (e.g. patient is already "diseased" entering the cohort).
- **Class 2** → Patient diagnosed with the specified disease earlier than the 'days before' window (e.g. patient is already "diseased" entering the cohort).
- $t_d < t_b$
- **Class 3** &rarr; Patient has not been monitored long enough to definitively rule-out disease.
- $t_h < t_a$
Expand Down