Reading specific lines (with some frequency) with textscan

Hello all,

I have an ascii file that I need to process. It is in this format:
"
 "Surface Integral Report"

           Area-Weighted Average
             Radiation Heat Flux (w/m2)
-------------------------------- --------------------
                            wall 34.189476
                         wall_aw 28.288052
                ---------------- --------------------
                             Net 34.208313

           Area-Weighted Average
             Radiation Heat Flux (w/m2)
-------------------------------- --------------------
                            wall 34.324566
                         wall_aw 28.363642
                ---------------- --------------------
                             Net 34.345776
"
and the file has more than 1000 of this sets but all I need is the "Net" values. How can I use textscan in a way that reads the file and only make an array of Net values?

Any help is appreciated.
Cheers,
Negin