Natural Language Understanding (NLU) algorithms in mPower analyze clinical report text to automatically identify key findings, follow-ups, and quality-related events. These algorithms turn unstructured narrative reports into actionable insights for analytics and quality review.
What NLU algorithms do
mPower's Clinical Language Understanding (CLU) engine powers NLU algorithms. They extract structured data from report text to support analytics and mPower QC workflows.
NLU algorithms help detect:
- Clinical measurements and lesion sizes
- Critical results documented in reports
- Follow‑up recommendations
- Documentation mismatches, such as:
- Laterality mismatches
- Ultrasound mismatches
- Sex mismatches
- Age mismatches
This automated extraction enables faster review and more consistent quality monitoring.
Where NLU results are used
NLU-driven data appears across mPower features, including:
- Measurement Finding dashboards
- Follow-up Recommendations dashboards
- Critical Results dashboards
- Mismatch dashboards
- Quality Measure dashboards
- mPower QC review workflows
These views rely on NLU to surface clinically relevant information without manual data entry.
When to reference NLU algorithms
Understand NLU algorithms when you want to:
- Interpret results shown in QC and analytics dashboards
- Explain why reports are flagged for review
- Identify patterns across follow-ups, critical results, or mismatches
- Support quality and performance discussions
Next up: Review the Follow‑up Recommendations dashboard
Track completed and overdue follow‑up recommendations to reduce missed follow‑ups and support timely patient care. Let's go