I used multiline from ssc to produce the following graph. Is possible to produce a table using data from the graph?
Graph.gph 
Code:
clear input int weight_birth1 byte volume_transfused float(volume_transfused1 total_hemoglobin total_hemocrit) 740 . 20 12.4 41.7 740 20 20 12.4 41.7 740 . 20 . . 880 20 20 7.3 21.4 880 . 20 17.1 49.4 880 . 20 . . 790 20 20 9.6 29.5 790 20 20 10.3 30.4 790 20 20 11.4 33.4 790 . 20 18.1 51.8 790 . 20 . . 790 20 16.666666 8.8 25.7 790 20 16.666666 10.4 29.7 790 . 16.666666 15.2 43.2 790 . 16.666666 . . 790 10 16.666666 . . 500 20 18.055555 7 20.2 500 20 18.055555 9.2 25.5 500 20 18.055555 9.4 27.2 500 20 18.055555 10.3 27.7 500 15 18.055555 9 27.8 500 20 18.055555 13.6 36.3 500 10 18.055555 11.1 36.4 500 . 18.055555 13.2 37.5 500 20 18.055555 13.2 37.5 500 20 18.055555 14.1 41.8 500 . 18.055555 . . 500 20 18.055555 . . 500 20 18.055555 . . 500 20 18.055555 . . 500 10 18.055555 . . 500 20 18.055555 . . 500 20 18.055555 . . 500 20 18.055555 . . 500 10 18.055555 . . 500 20 18.055555 . . 840 10 16.666666 14 44.1 840 . 16.666666 14 44.1 840 10 16.666666 . . 840 20 16.666666 . . 840 . 16.666666 . . 840 20 16.666666 . . 840 20 16.666666 . . 840 20 16.666666 . . 720 20 15 8.1 24.4 720 10 15 10.6 32.7 720 . 15 15.7 44.1 720 . 15 . . 1050 40 30 9.5 27.7 1050 . 30 14.1 44.8 1050 10 30 14.1 44.8 1050 40 30 . . 1050 40 30 . . 1050 20 30 . . 1050 . 30 . . 740 20 20 8.9 27.1 740 20 20 9.3 28.6 740 . 20 13.2 39.3 740 . 20 . . 740 20 20 . . end
Code:
multiline weight_birth1 volume_transfused volume_transfused1 total_hemoglobin total_hemocrit, xla(14(2)44) recast(connected) separate by(legend(off) t1title(NICU transfusion)) ms(O D T) lc(black green orange blue red) mc(black green orange blue red)