Anatomy Quiz Master
Overview
Comprehensive anatomy education tool that generates interactive quizzes covering gross anatomy, neuroanatomy, and clinical anatomy with adaptive difficulty and detailed explanations.
Key Capabilities:
- - Regional Quizzes: Head/neck, thorax, abdomen, pelvis, limbs
- Multiple Question Types: Identification, function, clinical correlation
- Adaptive Difficulty: Basic, intermediate, advanced levels
- Image Integration: Label identification with anatomical images
- Progress Tracking: Performance analytics and weak area identification
- Exam Mode: Timed simulations for USMLE-style preparation
When to Use
✅ Use this skill when:
- - Medical students preparing for anatomy practical exams
- Self-assessment after anatomy lectures or dissections
- Identifying weak anatomical regions for focused study
- Creating practice questions for study groups
- Remediation for students who failed anatomy assessments
- Preparing for USMLE Step 1 anatomy questions
- Teaching assistants generating quiz materials for labs
❌ Do NOT use when:
- - Primary learning resource for anatomy → Use textbooks/atlas first
- Substitute for cadaver lab attendance → Use for supplemental practice only
- Pathology or physiology questions → Use specialized skills for those topics
- Board exam registration or scheduling → Use official NBME resources
Integration:
- - Upstream:
usmle-case-generator (clinical context), anki-card-creator (flashcard export) - Downstream:
study-limitations-drafter (weakness analysis), performance-tracker (progress monitoring)
Core Capabilities
1. Regional Anatomy Quizzes
Generate focused quizzes by body region:
CODEBLOCK0
Supported Regions:
| Region | Subtopics | Question Types |
|---|
| Head & Neck | Skull, cranial nerves, triangles, viscera | Identification, pathways, clinical |
| Thorax |
Heart, lungs, mediastinum, pleura | Relations, auscultation, imaging |
|
Abdomen | GI tract, retroperitoneum, vessels | Peritoneal reflections, vascular supply |
|
Pelvis | Organs, perineum, walls | Gender differences, clinical correlations |
|
Upper Limb | Shoulder, arm, forearm, hand | Muscle actions, innervation, clinical |
|
Lower Limb | Hip, thigh, leg, foot | Gait, compartments, clinical exams |
|
Back | Vertebral column, spinal cord, muscles | Levels, landmarks, clinical |
2. Neuroanatomy Pathway Tracing
Specialized quizzes for neural pathways:
CODEBLOCK1
Pathway Types:
- - Motor Pathways: Corticospinal, corticobulbar, basal ganglia circuits
- Sensory Pathways: Dorsal column, spinothalamic, trigeminal
- Cranial Nerves: All 12 nerves with nuclei and clinical tests
- Reflex Arcs: Deep tendon, superficial, visceral
- Vascular: Arterial supply, venous drainage, stroke syndromes
3. Clinical Correlation Questions
Integrate anatomy with clinical scenarios:
CODEBLOCK2
Question Formats:
CODEBLOCK3
4. Adaptive Learning System
Adjust difficulty based on performance:
CODEBLOCK4
Adaptive Features:
- - Spaced Repetition: Re-test incorrect topics at optimal intervals
- Difficulty Scaling: Increase level after 3 consecutive correct answers
- Time Pressure: Gradually reduce time limits for speed practice
- Weakness Identification: Track performance by anatomical structure
Common Patterns
Pattern 1: Pre-Exam Comprehensive Review
Scenario: Student preparing for anatomy practical exam in 2 weeks.
CODEBLOCK5
Study Schedule:
- - Week 1: Comprehensive quizzes (all regions)
- Week 2: Focus on <80% score regions
- 3 days before: Timed practice exam
- Day before: Light review of marked difficult questions
Pattern 2: Lab Session Preparation
Scenario: Student preparing for cadaver lab on upper limb.
CODEBLOCK6
Lab Integration:
- - Pre-lab: 15-minute identification quiz
- During lab: Reference key landmarks
- Post-lab: Clinical correlation quiz linking anatomy to disease
Pattern 3: USMLE Step 1 Preparation
Scenario: Medical student preparing for USMLE Step 1.
CODEBLOCK7
USMLE Features:
- - Clinical vignette format
- Image-based questions (radiology, pathology)
- Two-step reasoning (identify structure → clinical implication)
- Time pressure simulation (60-90 seconds per question)
Pattern 4: Teaching Assistant Lab Quiz
Scenario: TA needs to generate weekly lab quizzes.
CODEBLOCK8
TA Tools:
- - Station-based practical exam format
- Answer keys with acceptable variations
- Grading rubrics
- Performance statistics by question
Complete Workflow Example
Comprehensive anatomy study session:
CODEBLOCK9
Python API:
CODEBLOCK10
Quality Checklist
Question Quality:
- - [ ] Anatomical accuracy verified against standard atlases (Netter, Gray's)
- [ ] Clinical correlations reviewed by licensed physicians
- [ ] Multiple difficulty levels appropriately calibrated
- [ ] Distractors (wrong answers) are plausible and educational
- [ ] Explications explain why correct answer is right
- [ ] Image quality sufficient for identification (resolution, labeling)
Educational Value:
- - [ ] Questions test high-yield anatomy (clinically relevant)
- [ ] Progressive difficulty builds knowledge systematically
- [ ] Clinical scenarios reflect real patient presentations
- [ ] Explanations include anatomical reasoning
Technical Quality:
- - [ ] Randomization prevents pattern recognition
- [ ] No duplicate questions in quiz banks
- [ ] Image files properly licensed or original
- [ ] Accessibility compliance (alt text for images)
Before Use:
- - [ ] CRITICAL: Faculty review for anatomical accuracy
- [ ] Pilot test with target student population
- [ ] Time limits appropriate for difficulty
- [ ] Answer key double-checked for errors
Common Pitfalls
Content Issues:
- - ❌ Outdated anatomical knowledge → Teaching old terminology
- ✅ Use current Terminologia Anatomica standards
- - ❌ Nit-picky details → Testing obscure structures rarely clinically relevant
- ✅ Focus on high-yield anatomy that appears in clinical practice
- - ❌ Unclear images → Poor resolution or confusing labels
- ✅ Use high-quality images; test label legibility at screen resolution
Educational Issues:
- - ❌ Questions too easy → No learning benefit
- ✅ Calibrate to student level; aim for 60-80% success rate
- - ❌ No clinical context → Pure memorization without application
- ✅ Include clinical correlation questions
- - ❌ Punitive difficulty → Discouraging rather than challenging
- ✅ Provide encouraging feedback; focus on improvement
Technical Issues:
- - ❌ Predictable patterns → Students game the system
- ✅ Randomize question order and distractor placement
- - ❌ No progress tracking → Can't identify weak areas
- ✅ Implement analytics to guide focused study
References
Available in references/ directory:
- -
netter_atlas_correlation.md - Question-to-atlas page mapping - INLINECODE6 - Standard anatomical terminology
- INLINECODE7 - NBME anatomy topic frequencies
- INLINECODE8 - High-yield clinical anatomy scenarios
- INLINECODE9 - Licensed anatomical image repositories
- INLINECODE10 - Bloom's taxonomy level alignment
Scripts
Located in scripts/ directory:
- -
main.py - CLI for quiz generation - INLINECODE13 - Core question generation engine
- INLINECODE14 - Specialized neuroanatomy questions
- INLINECODE15 - Clinical scenario integration
- INLINECODE16 - Personalized difficulty adjustment
- INLINECODE17 - Label identification with images
- INLINECODE18 - Performance analytics
- INLINECODE19 - Progress reports and statistics
Limitations
- - Cadaver Images: Cannot replace hands-on dissection experience
- 3D Spatial Relations: 2D images may not convey depth relationships
- Variability: Normal anatomical variation not fully captured
- Updates: Anatomical knowledge evolves; requires periodic review
- Cultural Sensitivity: Some anatomical terms may vary by region
- Disability Accommodation: Image-based questions need alternatives for visually impaired students
Parameters
| Parameter | Type | Default | Required | Description |
|---|
| INLINECODE20 , INLINECODE21 | string | upperlimb | No | Anatomical region (upperlimb, lowerlimb, thorax, abdomen, pelvis, headneck, neuroanatomy) |
| INLINECODE22 , INLINECODE23 |
string | intermediate | No | Difficulty level (basic, intermediate, advanced) |
|
--count,
-c | int | 1 | No | Number of questions to generate |
|
--output,
-o | string | - | No | Output file path (JSON format) |
|
--format | string | json | No | Output format (json or text) |
|
--list-regions | flag | - | No | List all available regions and exit |
Usage
Basic Usage
CODEBLOCK11
Risk Assessment
| Risk Indicator | Assessment | Level |
|---|
| Code Execution | Python script executed locally | Low |
| Network Access |
No external API calls | Low |
| File System Access | Read/Write to specified output files only | Low |
| Instruction Tampering | Standard prompt guidelines | Low |
| Data Exposure | Output saved only to specified location | Low |
Security Checklist
- - [x] No hardcoded credentials or API keys
- [x] No unauthorized file system access (../)
- [x] Output does not expose sensitive information
- [x] Prompt injection protections in place
- [x] Input validation for all parameters
- [x] Output directory restricted to workspace
- [x] Script execution in sandboxed environment
- [x] Error messages sanitized
Prerequisites
CODEBLOCK12
Evaluation Criteria
Success Metrics
- - [x] Successfully generates quiz questions
- [x] Supports multiple anatomical regions
- [x] Provides correct answers with explanations
- [x] Handles edge cases (invalid regions, etc.)
Test Cases
- 1. Basic Functionality: Generate single question → Returns valid question with options
- Edge Case: Invalid region → Graceful error message
- Multiple Questions: Generate 10 questions → Returns array of questions
Lifecycle Status
- - Current Stage: Draft
- Next Review Date: 2026-03-06
- Known Issues: None
- Planned Improvements:
- Add image support for visual identification
- Expand question bank
- Add performance analytics
🧠 Learning Tip: Anatomy is best learned through repeated exposure in multiple contexts. Use these quizzes to reinforce cadaver lab learning, not replace it. Focus on understanding relationships and clinical significance, not just memorization.
解剖学测验大师
概述
综合性解剖学教育工具,可生成涵盖大体解剖学、神经解剖学和临床解剖学的交互式测验,具有自适应难度和详细解释。
核心能力:
- - 区域测验:头颈部、胸部、腹部、骨盆、四肢
- 多种题型:识别、功能、临床关联
- 自适应难度:基础、中级、高级水平
- 图像整合:解剖图像标注识别
- 进度追踪:表现分析和薄弱区域识别
- 考试模式:USMLE风格限时模拟
使用时机
✅ 使用此技能的场景:
- - 医学生准备解剖学实践考试
- 解剖学讲座或解剖课后自我评估
- 识别薄弱解剖区域进行针对性学习
- 为学习小组创建练习题
- 为解剖学评估不及格的学生提供补救
- 准备USMLE第一步解剖学题目
- 助教为实验室生成测验材料
❌ 不应使用的场景:
- - 作为解剖学主要学习资源 → 应首先使用教科书/图谱
- 替代尸体解剖实验室出勤 → 仅用于补充练习
- 病理学或生理学问题 → 使用这些主题的专业技能
- 委员会考试注册或安排 → 使用官方NBME资源
集成:
- - 上游:usmle-case-generator(临床情境),anki-card-creator(闪卡导出)
- 下游:study-limitations-drafter(弱点分析),performance-tracker(进度监控)
核心能力
1. 区域解剖学测验
按身体区域生成针对性测验:
python
from scripts.quiz_generator import QuizGenerator
generator = QuizGenerator()
生成胸部测验
quiz = generator.generate_quiz(
region=thorax,
topics=[heart, lungs, mediastinum, thoracic_wall],
difficulty=intermediate,
n_questions=20
)
导出至LMS
quiz.export(format=json, filename=thorax_quiz.json)
支持的区域:
| 区域 | 子主题 | 题型 |
|---|
| 头颈部 | 颅骨、脑神经、三角区、内脏 | 识别、通路、临床 |
| 胸部 |
心脏、肺、纵隔、胸膜 | 关系、听诊、影像 |
|
腹部 | 胃肠道、腹膜后腔、血管 | 腹膜反折、血管供应 |
|
骨盆 | 器官、会阴、壁 | 性别差异、临床关联 |
|
上肢 | 肩部、上臂、前臂、手 | 肌肉动作、神经支配、临床 |
|
下肢 | 髋部、大腿、小腿、足 | 步态、筋膜室、临床检查 |
|
背部 | 脊柱、脊髓、肌肉 | 节段、标志、临床 |
2. 神经解剖学通路追踪
神经通路的专业测验:
python
神经解剖学测验
neuro
quiz = generator.generateneuro_quiz(
pathway
type=motor, # 或 sensory, cranialnerves, reflexes
include_lesions=True,
clinical_correlations=True
)
通路类型:
- - 运动通路:皮质脊髓束、皮质延髓束、基底节环路
- 感觉通路:背柱、脊髓丘脑束、三叉神经
- 脑神经:全部12对脑神经及其核团和临床检查
- 反射弧:深腱反射、浅反射、内脏反射
- 血管:动脉供应、静脉引流、卒中综合征
3. 临床关联题
将解剖学与临床场景整合:
python
clinicalquiz = generator.generateclinical_quiz(
region=abdomen,
scenariotypes=[surgery, radiology, physicalexam],
difficulty=advanced
)
题目格式:
临床场景:
一名45岁男性,表现为向背部放射的上腹痛。
CT显示小网膜囊内有一肿块。
问题:哪条动脉紧贴胰腺体部后方走行,在切除术中可能面临风险?
A) 脾动脉
B) 肠系膜上动脉
C) 肝总动脉
D) 胃左动脉
正确答案:B) 肠系膜上动脉
解释:SMA自L1水平从主动脉发出,经过胰腺颈部后方和钩突前方...
4. 自适应学习系统
根据表现调整难度:
python
from scripts.adaptive import AdaptiveEngine
engine = AdaptiveEngine()
追踪学生表现
student
progress = engine.trackperformance(
student
id=student001,
quiz_results=results,
time
perquestion=True
)
生成针对薄弱区域的个性化测验
personalized = engine.generate
adaptivequiz(
student
progress=studentprogress,
focus
areas=[thoraxvessels, cranial_nerves],
mastery_threshold=0.80
)
自适应功能:
- - 间隔重复:以最佳间隔重新测试错误主题
- 难度递进:连续答对3题后提升难度
- 时间压力:逐步减少时间限制以训练速度
- 弱点识别:按解剖结构追踪表现
常见模式
模式1:考前全面复习
场景:学生准备2周后的解剖学实践考试。
bash
生成全身综合测验
python scripts/main.py \
--mode comprehensive \
--regions all \
--difficulty intermediate \
--n-questions 100 \
--timed \
--output pre
practiceexam.json
针对识别的薄弱区域
python scripts/main.py \
--mode adaptive \
--focus abdomen,pelvis \
--difficulty advanced \
--n-questions 30 \
--output weak
areasreview.json
学习计划:
- - 第1周:综合测验(所有区域)
- 第2周:重点复习得分<80%的区域
- 考前3天:限时模拟考试
- 考前1天:轻量复习标记的难题
模式2:实验室课程准备
场景:学生准备上肢尸体解剖实验室。
python
实验前识别测验
pre
lab = generator.generateimage_quiz(
region=upper_limb,
structure_types=[muscles, vessels, nerves],
label_type=pins, # 标注识别格式
n_questions=15
)
实验后临床关联
post
labclinical = generator.generate
clinicalquiz(
region=upper_limb,
clinical
types=[fractures, nerveinjuries, vascular]
)
实验室整合:
- - 实验前:15分钟识别测验
- 实验期间:参考关键标志
- 实验后:将解剖学与疾病关联的临床关联测验
模式3:USMLE第一步准备
场景:医学生准备USMLE第一步。
bash
USMLE风格临床解剖学
python scripts/main.py \
--mode usmle \
--clinical-focus \
--mix-basic-advanced 70:30 \
--n-questions 40 \
--timed-per-question 60 \
--output usmle
anatomypractice.json
USMLE功能:
- - 临床病例格式
- 基于图像的题目(放射学、病理学)
- 两步推理(识别结构→临床意义)
- 时间压力模拟(每题60-90秒)
模式4:助教实验室测验
场景:助教需要生成每周实验室测验。
python
每周实验室测验
ta
quiz = generator.generateta_quiz(
week_number=5,
region=thorax,
practical_stations=8,
time
perstation=3, # 分钟
include
prosectionimages=True
)
自动生成答案
answer
key = taquiz.generate
answerkey(
include
acceptablevariations=True,
grading
rubric=partialcredit
)
助教工具:
- - 基于站点的实践考试格式
- 包含可接受变体的答案
- 评分标准
- 按题目统计的表现数据
完整工作流程示例
综合解剖学学习课程:
bash
步骤1:诊断测验以识别薄弱区域
python scripts/main.py \
--mode diagnostic \
--regions all \
--n-questions 50 \
--output diagnostic_results.json
步骤2:生成针对性学习计划
python scripts/main.py \
--analyze-results diagnostic_results.json \
--generate-study-plan \
--days 14 \
--output study_plan.md
步骤3:按计划进行每日测验
python scripts/main.py \
--mode daily \
--study-plan study_plan.md \
--day 1 \
--output day1_quiz.json
步骤4:间隔重复复习
python scripts/main.py \