Count number number of rows until value is reached and do this within specific ranged within a dataset

I'm a beginner R user (not entirely sure if it's the most appropriate language for this use) and am having difficulty coming up with a solution for what I think is probably relatively easy.

I have a dataset that looks as follows:

    Filename    Label   seg_Start   seg_End t_ms    CPP
1   BJB1NRL.mat V       231.537     444.039 230     0.000
2   BJB1NRL.mat V       231.537     444.039 240     0.000
3   BJB1NRL.mat V       231.537     444.039 250     19.553
4   BJB1NRL.mat V       231.537     444.039 260     16.063
5   BJB1NRL.mat V       231.537     444.039 270     18.090
6   BJB1NRL.mat V       231.537     444.039 280     21.547

The fifth column (t_ms) is time in milliseconds, and the third and fourth columns are also time. Once fifth column reaches the value in seg_End (rounded), the third and fourth column change, i.e:

57  BJB1NRL.mat V   554.141     887.246     890     14.125
58  BJB1NRL.mat V   1081.457    1269.050    1080    0.000
59  BJB1NRL.mat V   1081.457    1269.050    1090    0.000
60  BJB1NRL.mat V   1081.457    1269.050    1100    22.398
61  BJB1NRL.mat V   1081.457    1269.050    1110    20.006
62  BJB1NRL.mat V   1081.457    1269.050    1120    17.931

What I have to do is straightfoward to do manually, but I'm struggling to create a script to do it for me. For each segment (i.e. 231-444, then 1081-1269 etc) I need to see how many milliseconds it takes for the value in the final column to reach a threshold value (e.g. 28). That is, I need to count how many rows from the segment starting (t_ms = seg_Start) until the value in the final row reaches this threshold value, since each row is 10ms. If the value never reaches the threshold then a value of 0 should be given.

So far, my code is mostly my attempts to split the dataset up accordingly to be analysed. It is likely inefficient, and more worringly, incorrect.

The dataset has hundreds of different 'filenames' which need to be analysed separately, so I split those up first.

Result <- #The dataset, currently a .csv file
filenames <- unique(Result$Filename)
levels <- nlevels(filenames)

for(i in 1:levels){
  nam <- paste("Sub", i, sep = "")
  assign(nam, subset(Result, Filename == filenames[i]))

  }

Sub <- lapply(ls(pattern="Sub[0-9]+"), function(x) get(x))

# Create list with segment end times
i <- NULL
for(i in 1:levels){
  nam <- paste("unique.Sub", i, sep = "")
  assign(nam, unique(Sub[[i]]$seg_End))
}

unique.Sub <- lapply(ls(pattern="unique.Sub[0-9]+"), function(x) get(x))

# Round these values to nearest 10
i <- NULL
for(i in 1:levels){
  unique.Sub[[i]] <- round(unique.Sub[[i]], digits=-1)
}

# Create list with segment start times
i <- NULL
for(i in 1:levels){
  nam <- paste("unique2.Sub", i, sep = "")
  assign(nam, unique(Sub[[i]]$seg_Start))
}

unique2.Sub <- lapply(ls(pattern="unique2.Sub[0-9]+"), function(x) get(x))

i <- NULL
for(i in 1:levels){
  unique2.Sub[[i]] <- round(unique2.Sub[[i]], digits=-1)
}

# Rename variables for clarity
segStart <- sapply(unique2.Sub, function(x) x);
segEnd <- sapply(unique.Sub, function(x) x)`s

I then think I have to go on to the actual analysis, something like

lapply(Sub, function(i){
  if Sub$t_ms < segStart && < segEnd
  which.max(Sub$t_ms >= 26)

I hope what I am aiming for makes sense, even if the code I've come up with so far does not. As I said, my impression is that there is an easier way to do the analysis than I think. Any help is hugely appreciated.