triage/analysis/dataset_statistics.py

#!/usr/bin/env python3
"""
统计dataset/bbb.json中所有病例的一级科室和二级科室集合
"""

import json
from pathlib import Path
from collections import Counter


def load_dataset(file_path: str) -> list:
    """加载数据集"""
    try:
        with open(file_path, 'r', encoding='utf-8') as f:
            return json.load(f)
    except Exception as e:
        print(f"加载文件 {file_path} 时出错: {e}")
        return []


def analyze_departments(data: list) -> dict:
    """分析科室分布"""
    
    # 收集一级科室和二级科室
    level1_departments = []
    level2_departments = []
    
    # 建立一级到二级的映射关系
    level1_to_level2 = {}
    
    for case in data:
        level1 = case.get('一级科室', '').strip()
        level2 = case.get('二级科室', '').strip()
        
        if level1:
            level1_departments.append(level1)
            
        if level2:
            level2_departments.append(level2)
            
        # 建立映射关系
        if level1 and level2:
            if level1 not in level1_to_level2:
                level1_to_level2[level1] = set()
            level1_to_level2[level1].add(level2)
    
    # 统计计数
    level1_counter = Counter(level1_departments)
    level2_counter = Counter(level2_departments)
    
    return {
        'level1_counter': level1_counter,
        'level2_counter': level2_counter,
        'level1_to_level2': {k: list(v) for k, v in level1_to_level2.items()},
        'total_cases': len(data),
        'unique_level1': len(set(level1_departments)),
        'unique_level2': len(set(level2_departments))
    }


def print_statistics(stats: dict):
    """打印统计结果"""
    
    print("=" * 60)
    print("DATASET 科室统计报告")
    print("=" * 60)
    
    print(f"总病例数: {stats['total_cases']}")
    print(f"一级科室种类数: {stats['unique_level1']}")
    print(f"二级科室种类数: {stats['unique_level2']}")
    print()
    
    print("一级科室分布:")
    print("-" * 40)
    for dept, count in sorted(stats['level1_counter'].items(), key=lambda x: x[1], reverse=True):
        percentage = (count / stats['total_cases']) * 100
        print(f"  {dept}: {count} 例 ({percentage:.1f}%)")
    
    print()
    print("二级科室分布:")
    print("-" * 40)
    for dept, count in sorted(stats['level2_counter'].items(), key=lambda x: x[1], reverse=True):
        percentage = (count / stats['total_cases']) * 100
        print(f"  {dept}: {count} 例 ({percentage:.1f}%)")
    
    print()
    print("一级科室 → 二级科室映射:")
    print("-" * 40)
    for level1, level2_list in sorted(stats['level1_to_level2'].items()):
        print(f"  {level1}:")
        for level2 in sorted(level2_list):
            count = stats['level2_counter'][level2]
            print(f"    - {level2}: {count} 例")
        print()


def save_statistics(stats: dict, output_file: str):
    """保存统计结果"""
    
    # 准备保存的数据
    save_data = {
        '一级科室列表': sorted(list(stats['level1_counter'].keys())),
        '二级科室列表': sorted(list(set(stats['level2_counter'].keys()))),
        '一级科室计数': dict(stats['level1_counter']),
        '二级科室计数': dict(stats['level2_counter']),
        '一级科室到二级科室映射': stats['level1_to_level2'],
        '统计信息': {
            '总病例数': stats['total_cases'],
            '一级科室种类数': stats['unique_level1'],
            '二级科室种类数': stats['unique_level2']
        }
    }
    
    with open(output_file, 'w', encoding='utf-8') as f:
        json.dump(save_data, f, ensure_ascii=False, indent=2)


def main():
    """主函数"""
    
    # 设置路径
    dataset_file = Path(__file__).parent.parent / "dataset" / "bbb.json"
    output_file = Path(__file__).parent.parent / "analysis" / "dataset_department_statistics.json"
    
    print(f"正在加载数据集: {dataset_file}")
    data = load_dataset(str(dataset_file))
    
    if not data:
        print("无法加载数据集")
        return
    
    print(f"成功加载 {len(data)} 个病例")
    
    stats = analyze_departments(data)
    print_statistics(stats)
    
    save_statistics(stats, str(output_file))
    print(f"统计结果已保存到: {output_file}")
    
    # 额外输出纯列表格式
    print("\n" + "=" * 60)
    print("科室列表（纯文本格式）")
    print("=" * 60)
    
    print("一级科室集合:")
    for dept in sorted(list(stats['level1_counter'].keys())):
        print(f"  '{dept}'")
    
    print("\n二级科室集合:")
    for dept in sorted(list(set(stats['level2_counter'].keys()))):
        print(f"  '{dept}'")


if __name__ == "__main__":
    main()