Structured Reporting

Structured Reporting

You are an expert in structured radiology reporting. Your role is to help create, convert, and optimize standardized radiology reports.

Reporting Standards

By Modality/Body Part

| Standard | Modality | Use Case | |----------|----------|----------| | BI-RADS | Mammography | Breast imaging assessment | | LI-RADS | MRI/CT | Liver imaging reporting | | PI-RADS | MRI | Prostate imaging | | TI-RADS | Ultrasound | Thyroid imaging | | Lung-RADS | CT | Lung cancer screening | | RADS | Various | Multiple standardized systems |

BI-RADS Templates

BI-RADS Assessment Categories

| Category | Assessment | Management | |---------|------------|------------| | 0 | Incomplete | Additional imaging | | 1 | Negative | Routine screening | | 2 | Benign | Routine screening | | 3 | Probably Benign | Short interval follow-up | | 4A | Suspicious | Biopsy | | 4B | Suspicious | Biopsy | | 4C | Suspicious | Biopsy | | 5 | Highly Suspicious | Biopsy | | 6 | Known Malignancy | Treatment |

BI-RADS Report Template

## MAMMOGRAPHY REPORT

**Patient:** [Name] | **DOB:** [Date] | **Accession:** [Number]

### Clinical History:
[History]

### Technique:
[Standard views, supplementary views]

### Comparison:
[Prior studies compared]

### Findings:

**Left Breast:**
- [Location]: [Finding] [size if applicable]

**Right Breast:**
- [Location]: [Finding] [size if applicable]

### Assessment:
**BI-RADS: [Category]** - [Description]

### Recommendations:
[Follow-up recommendations]

LI-RADS Templates

LI-RADS Categories

| Category | Definition | |----------|------------| | LR-1 | Definitely benign | | LR-2 | Probably benign | | LR-3 | Intermediate probability of malignancy | | LR-4 | Probably HCC | | LR-5 | Definitely HCC | | LR-M | Probably malignant, not HCC specific |

LI-RADS Report Template

## LIVER IMAGING REPORT

**Study:** MRI Liver with contrast | **Accession:** [Number]

### Clinical History:
[ cirrhosis surveillance / elevated AFP / etc. ]

### Technique:
[ MRI sequences, contrast agent ]

### Liver Observations:

| # | Size (mm) | Segment | Category | APHE | Washout | Capsule | Threshold Growth |
|---|-----------|--------|----------|------|---------|--------|-----------------|
| 1 | 25 | S8 | LR-5 | Yes | Yes | Yes | - |

### Additional Findings:
[Other benign findings, etc.]

### Impression:
1. [LR-X] - [Size] hemodynamically [characteristic] in [Location]

### Recommendations:
[Follow-up imaging / biopsy / treatment]

PI-RADS Templates

PI-RADS Assessment

| Category | Assessment | |----------|------------| | PI-RADS 1 | Very low | | PI-RADS 2 | Low | | PI-RADS 3 | Intermediate | | PI-RADS 4 | High | | PI-RADS 5 | Very high |

PI-RADS Report Template

## PROSTATE MRI REPORT

**Study:** Multiparametric MRI Prostate | **Accession:** [Number]

### Clinical History:
[PSA level, previous biopsies, clinical suspicion]

### Technique:
[T2W, DWI, DCE sequences]

### Prostate Volume:
[Volume in cc]

### Findings:

**Transition Zone:**
- [Location, size, PIRADS score, description]

**Peripheral Zone:**
- [Location, size, PIRADS score, description]

**Extra-prostatic Extension:**
[Present / Not identified]

**Seminal Vesicles:**
[Involved / Not involved]

**Lymph Nodes:**
[Suspicious / Not suspicious]

### Assessment:
**PI-RADS: [Score]** - [Clinical significance]

### Recommendations:
[Biopsy guidance / follow-up]

TI-RADS Templates

TI-RADS Categories

| Category | Risk | Recommendation | |----------|------|----------------| | TR1 | None | None | | TR2 | Very low | None | | TR3 | Low | Follow-up | | TR4 | Intermediate | FNA if criteria met | | TR5 | High | FNA recommended |

TI-RADS Report Template

## THYROID ULTRASOUND REPORT

**Study:** Thyroid Ultrasound | **Accession:** [Number]

### Clinical History:
[Indication for study]

### Findings:

**Right Lobe:**
- [Size], [echogenicity], [margins], [calcifications], [TI-RADS: TR-X]

**Left Lobe:**
- [Same parameters]

**Isthmus:**
- [Findings]

**Central Neck:**
- [Lymph nodes, if visible]

### Assessment:
**TI-RADS: [Category]** - [Description]

### Recommendations:
[FNA if indicated, follow-up schedule]

Lung-RADS Templates

Lung-RADS Categories

| Category | Findings | Management | |----------|----------|------------| | 0 | Incomplete | Additional evaluation | | 1 | No nodules | Continue annual screening | | 2 | Benign | Continue annual screening | | 3 | Probably benign | 6-month follow-up | | 4A | Suspicious | 3-month follow-up or PET | | 4B | Very suspicious | Chest CT with/without PET | | 4C | - | Evaluate as 4B | | S | Significant incidental finding | - |

Lung-RADS Report Template

## CT CHEST LUNG CANCER SCREENING REPORT

**Study:** Low-dose CT Chest | **Accession:** [Number]

### Clinical History:
[Age, smoking history, eligibility criteria]

### Findings:

**Lungs:**
- [Nodules identified with characteristics]
  - Location: [Lobe]
  - Size: [mm]
  - Type: [Solid/GGS/Mixed]
  - Margins: [Smooth/irregular/spiculated]

**Nodule Analysis:**
| Nodule | Location | Size | Type | Lung-RADS |
|--------|----------|------|------|-----------|
| 1 | RLL | 6mm | Solid | 2 |

### Assessment:
**Lung-RADS: [Category]**

### Recommendations:
[Based on category - follow-up/CT/PET/biopsy]

### Additional Findings:
[Any significant incidental findings]

Free-Text to Structured Conversion

def convert_free_to_structured(free_text_report):
    """Convert free-text report to structured format."""
    
    # Extract key sections
    sections = {
        "clinical_history": extract_section(free_text_report, "CLINICAL|HISTORY|INDICATION"),
        "findings": extract_findings(free_text_report),
        "impression": extract_section(free_text_report, "IMPRESSION|CONCLUSION"),
    }
    
    # Identify modality and body part
    modality = detect_modality(free_text_report)
    body_part = detect_body_part(free_text_report)
    
    # Select appropriate template
    if "BREAST" in free_text_report.upper():
        template = "BI-RADS"
    elif "LIVER" in free_text_report.upper():
        template = "LI-RADS"
    elif "PROSTATE" in free_text_report.upper():
        template = "PI-RADS"
    elif "THYROID" in free_text_report.upper():
        template = "TI-RADS"
    else:
        template = "GENERAL"
    
    return {
        "original": free_text_report,
        "structured": sections,
        "template": template,
        "modality": modality,
        "body_part": body_part
    }

Report Quality Checklist

• [ ] All required sections present
• [ ] Clinical indication documented
• [ ] Comparison with priors noted
• [ ] All findings described with size/characteristics
• [ ] Appropriate standard scoring used (BI-RADS, LI-RADS, etc.)
• [ ] Impression clearly states most significant findings
• [ ] Recommendations actionable and specific
• [ ] Incidental findings addressed

Related Skills

• radiology-report-analysis: For analyzing existing reports
• impression-generation: For AI-assisted impression writing
• patient-results-letter: For patient communication
• guideline-integration: For standard compliance

Examples

Example 1: BI-RADS Report

Generate a BI-RADS 4A report for a spiculated mass

**Left Breast:**
- 2 o'clock: 15mm spiculated mass with suspicious calcifications

**Assessment:**
BI-RADS: 4A - Low suspicion for malignancy

**Recommendations:**
Ultrasound-guided biopsy recommended

Example 2: Free-Text to Structured

Convert: "CT chest shows 2.5cm mass in right upper lobe. No mediastinal lymphadenopathy. Suspicion for malignancy."

**Findings:**
- Lungs: 25mm spiculated mass, right upper lobe
- Mediastinum: No lymphadenopathy

**Assessment:**
- Lung mass, concerning for malignancy

**Recommendations:**
- PET/CT for staging
- CT-guided biopsy

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: aizech
• Source: aizech/clinical-skills
• License: MIT
• Homepage: https://www.corpusanalytica.com/skills/

Install and usage instructions live in the source repository linked above.