正在尝试自定义此正则表达式以提取医疗报告结果

Question

尝试更正此正则表达式Lung[ ]+Rads[ ]+Catagory[ ]+(\d+[A-Z]*)从非结构化放射学报告中的Lung Rads分类中提取1、1S、2、2S、3、3S、4A、4AS、4B、4BS、4X、4XS以及C对这些倍数的潜在附加值，其中的语言可能看起来像：

LUNG RADS CATEGORY 2
Lung RADS category 1, 
Lung RADS category 2. 
LUNG RADS CATEGORY 1
Lung RADS 3S.
Lung RADS Category 1:
Lung RADS Category 1 (S):
LUNG RADS CATEGORY 1 S
Lung RADS category: 2S.
Lung RADS: 2C
Lung RADS category 4B,
Lung RADS category 1S.
Lung RADS: 3.
Lung RADS category I
Lung RADS 2
LUNG RADS CATEGORY:I
LUNGRADS 2
LUNGRAD 2
LUNG-RAD 3

Answer 1

您可以使用以下正则表达式来匹配感兴趣的字符串部分。

(?<=^(?:(?:Lung|LUNG) RADS:?(?: CATEGORY:?| [cC]ategory:?)? ?|LUNGRADS? |LUNG-RAD ))(?:d[A-Z]?|[A-Z])(?![A-Z])

Javascript演示

Javascript的正则表达式引擎执行以下操作。

(?<=                      begin positive lookbehind
^                       match beginning of line
(?:                     begin a non-cap grp
(?:Lung|LUNG) RADS:?  match 'Lung' or 'LUNG' followed by a a space,
then 'RADS', opt followed by ':
(?:                   begin non-cap grp
[ ]CATEGORY:?       match space, then 'CATEGORY', then opt ':"
|                   or
[ ][cC]ategory:?    match space then 'category' or 'Category'
then opt ':'
)                     end non-cap grp
?                     opt match non-cap grp
[ ]?                  opt match a space
|                     or
LUNGRADS?[ ]          match 'LUNGRAD ' or 'LUNGRADS '
|                     or
LUNG-RAD[ ]           match 'LUNG-RAD '       
)                       end non-cap grp
)                         end positive lookbehind
(?:                       begin non-cap grp
d[A-Z]?                match digit then opt cap letter
|                       or
[A-Z]                   match one cap letter
)                         end non-cap grp
(?![A-Z])                 match a cap letter in a negative lookahead