My study has 2 components:

1)

I have 10 organ and I want to find how accurate surgeons are using 2 seperate measurement techniques for measuring an angle on the organ compared to a gold standard.

4 surgeons each generate 1 measurement per organ per technique.

I want to pool the surgeons data together, making the asumption that they are representive, and then get some sort of value on the accuracy of the 2 techniques compared to gold standard.

Would a Lin's Concordance Correlation Coefficient for each technique be appropriate? Would that tell me how "good" each technique is in some sense?

Then, would it be appropriate to run ICC to see what the inter-observer agreement is for each of the 2 measurement techniques?

2)

I also have 1010 organs that were operated on by each surgeon 3 times using 3 different techniques.

After each "surgery" I measured the change in angle from pre to post surgery (if surgery was perfect, change would be 0)

I want to find some way of estimating the "accuracy" or "effectiveness" in some way of each of the 3 surgical techiques, pooling the data from the 4 surgeons.

could I do some sort of regression analysis to see overall how each technique performed in terms of having the smallest overall change in angle?

Could I also run an ICC to get a sense of the inter-observer variability for each surgical technique?

Thanks for reading