ALS-specific clinical content by category
Each row aggregates patients and records from all regex patterns within one clinical-content group. ALSFRS-R covers the functional rating scale and its four subdomain subscales; pulmonary captures FVC % predicted; diagnosis covers El Escorial / Awaji-Shima certainty and onset region; family history covers negative and positive constructs and genetic-mutation mentions; ECAS / FTD covers cognitive screening scores and FTD-spectrum references; treatment milestones cover dated procedures and drug starts (PEG, tracheostomy, NIV, riluzole, edaravone).
{_render_categories(summary)}All patterns
One row per regex pattern in the shipped library. Filter by name or category.
{_render_patterns(summary)}Patient × pattern matrix
PT-NNNN pseudonyms. Cell shows the number of records for that patient × that pattern; · = none. Patterns with fewer than k={k} unique patients are suppressed.
{_render_matrix(summary)}Captured-value distribution
For numeric patterns (ALSFRS-R scores, FVC % predicted), shows min / median / max across captured values. For categorical / freetext patterns (El Escorial certainty, onset region, family-history constructs), shows the top distinct values and their record counts. Numeric extractions are not de-duplicated per patient at this view; use the Patient × pattern matrix to scope to per-patient counts.
{_render_values(summary)}Source-record snippets
Up to 3 representative text excerpts per regex pattern, prioritizing unique-patient diversity. Snippet text truncated to 200 chars. Useful for source verification and pattern-tuning review.
{_render_snippets(snippet_groups)}About this dashboard
This dashboard surfaces ALS-specific clinical content captured by
note_extraction.py from the unstructured narrative
sections of the EHR feed (CCDA section narratives and decoded
documents — RTF, HTML, PDF). Unlike the device and exposure
dashboards, these are clinical findings (functional rating scores,
cognitive scores, diagnostic certainty, treatment milestones, family
history, genetic mutations) rather than risk factors or equipment.
Scope and limitations. The patterns shipped with
note_extraction.py are seed patterns calibrated against
a single registry (ARC). Adopters should validate every pattern
against their own narrative corpus before using captured values for
analysis; site-specific phrasing conventions, vendor-specific
template structures, and individual-clinician dictation patterns
vary substantially. Treat this dashboard as a chart-review-preparation
and pattern-tuning aid; do not admit numeric values (ALSFRS-R, FVC%,
ECAS) into downstream analysis without validation against the
patient's structured measurement record or the original note.
Privacy controls (baked in):
- Patient identifiers replaced with PT-NNNN pseudonyms, stable within this run.
- Snippet text truncated to 200 characters.
- Captured values truncated to 60 characters.
- Patterns with fewer than k={k} unique patients are suppressed entirely.
- Resource UUIDs are never emitted.
The companion module
note_extraction.py produces the underlying
note_extractions.csv; this dashboard is the privacy-safe
visualization layer over that CSV.