SYSTEMS AND METHODS FOR TRADE-OFF VISUAL ANALYSIS
A method for trial design analysis that includes receiving, for each trial design of a plurality of trial designs, a set of simulated performance criteria for a set of trial designs, and visualizing, on a graph, values for a first simulated performance criteria and a second simulated performance criteria from the set of simulated performance criteria for each trial design using a location of points on the graph corresponding to the set of trial designs. The method further includes identifying optimal designs based on an optimality criteria using the set of the simulated performance criteria, and determining a tradeoff metric for the first simulated performance criteria and the second simulated performance criteria. The method further includes displaying, the tradeoff metric as a set of lines on the graph, wherein a slope of the lines corresponds to a value of the tradeoff metric.
1 . A method comprising:
interpreting design data corresponding to a plurality of designs;
generating visualization data structured to show a design interface for comparing the plurality of designs; and
transmitting the visualization data.
2 . The method of claim 1 further comprising:
displaying the design interface;
interpreting a user input to the design interface; and
adjusting, in response to the user input, the design interface.
3 . The method of claim 1 , wherein the design interface comprises a 3-plot.
4 . The method of claim 3 , further comprising:
changing a first display element of a first graph of the 3-plot responsive to a corresponding change in a second display element of a second graph of the 3-plot.
5 . The method of claim 4 , wherein the first display element comprises a scale of the first graph.
6 . The method of claim 4 , wherein the first display element comprises a zoom level of the first graph.
7 . The method of claim 1 further comprising:
generating a line that connects a first point corresponding to a design on a first graph of a 3-plot with a second point of a second graph of the 3-plot and a third point of a third graph of the 3-plot, wherein the second point and the third point correspond to the design.
8 . The method of claim 1 , wherein the design interface comprises a heatmap.
9 . The method of claim 1 , wherein the design interface comprises a timeline.
10 . The method of claim 1 , further comprising:
identifying a plurality of design variables;
ordering values of the plurality of design variables according to a preference;
determining a level value to each value of the plurality of design variables according to the ordering;
generating a stratification geometry using the determined level values; and
comparing designs using the stratification geometry.
11 . The method of claim 10 , wherein comparing the designs comprises determining a Manhattan distance between designs using the stratification geometry.
12 . The method of claim 10 , wherein the level values are whole numbers.
13 . The method of claim 10 , wherein the plurality of design variables are a surrogate for qualitative criteria.
14 . The method of claim 10 , further comprising generating prose language comparing designs.
15 . The method of claim 10 , where comparing designs comprises comparing designs that are twins.
16 . The method of claim 10 , wherein comparing designs comprises determining Lexcode for each design based on the stratification geometry.
17 . The method of claim 10 , wherein the plurality of design variables includes at least three design variables.
18 . The method of claim 10 , further comprises identifying a plurality of designs based on quantitative criteria and comparing the designs using the stratification geometry.
19 . The method of claim 10 , further comprises identifying a plurality of designs based on quantitative criteria and comparing the designs using a Manhattan distance between the designs in the stratification geometry.
20 . The method of claim 10 , further comprises identifying a plurality of designs based on quantitative criteria and comparing the designs using a Lexcode of the designs in the stratification geometry.
21 . The method of claim 1 , further comprising:
receiving, for each trial design of a plurality of trial designs, a set of simulated performance criteria for a set of trial designs;
visualizing, on a graph, values for a first simulated performance criteria and a second simulated performance criteria from the set of simulated performance criteria for each trial design using a location of points on the graph corresponding to the set of trial designs;
identifying optimal designs based on an optimality criteria using the set of the simulated performance criteria;
determining a tradeoff metric for the first simulated performance criteria and the second simulated performance criteria; and
displaying, the tradeoff metric as a set of lines on the graph, wherein a slope of the lines corresponds to a value of the tradeoff metric.
22 . The method of claim 21 , wherein the lines include at least one Pareto optimal design.
23 . The method of claim 21 , wherein the lines include at least one convex hull design.
24 . The method of claim 21 , wherein a slope of one of the lines is configurable.
25 . The method of claim 21 further comprising:
selecting the tradeoff metric via a user input value.
26 . The method of claim 25 further comprising:
determining selected tradeoffs based at least in part on the selected tradeoff metric; and
identifying optimal designs based at least in part on the selected tradeoffs.
27 . The method of claim 21 further comprising:
highlighting optimal designs in response to changes in a slope of at least one of the set lines.
28 . The method of claim 21 , wherein the graph is two-dimensional.
29 . The method of claim 21 , wherein the graph is three-dimensional.