The following is an example of using the ClinicalTemplateReder for automated planning. Start by generated a Stand-alone executable with the Eclipse Script Wizard
Change your target framework to .NET 4.5.2. Go to the Nuget Package Manager and find the ClinicalTemplateReader package. You may find this by searching ESAPI.
Add the following using to your code to resolve the ClinicalTemplateReader related references:
using ClinicalTemplateReader;
In the Execute Method post the following code
static void Execute(Application app)
{
//Checking clinical templates
Console.WriteLine("Reading Clinical Templates");
string server = "Varian Image Server"; // <== update to your image server address e.g. "localhost", "10.11.12.13"
var clinicalProtols = new ClinicalTemplate(server);
var planStats_cp = clinicalProtols.GetPlanTemplateApprovals(true);
Console.WriteLine("Plan Template statistics with clinical protocols");
foreach (var planstat in planStats_cp)
{
Console.WriteLine($"\t{planstat.ApprovalStatus} - {planstat.Count}");
}
var objStats_cp = clinicalProtols.GetObjectiveTemplateApprovals(true);
Console.WriteLine("Objective Template Statistics with clinical protocols");
foreach (var objstat in objStats_cp)
{
Console.WriteLine($"\t{objstat.ApprovalStatus} - {objstat.Count}");
}
Console.ReadLine();
}
Running this should give the following response in the console.
The following code will allow you to select a plan template.
static void Execute(Application app)
{
// Checking clinical templates
string server = "Varian Image Server"; // <== update to your image server address e.g. "localhost", "10.11.12.13"
var clinicalProtols = new ClinicalTemplate(server);
var approvedTemplates = clinicalProtols.PlanTemplates.Where(x => x.Preview.ApprovalStatus.Contains("Approved")).ToList();
for (int i = 0; i < approvedTemplates.Count(); i++)
{
Console.WriteLine($"[{i}]. {approvedTemplates.ElementAt(i).Preview.ID} - {approvedTemplates.ElementAt(i).Preview.LastModified}");
}
Console.WriteLine($"\n\nPlease pick your plan template from the list (0 - {approvedTemplates.Count() - 1}):");
var plantemplatenumber = Convert.ToInt32(Console.ReadLine());
var plantemplate = approvedTemplates.ElementAt(plantemplatenumber);
Console.WriteLine("\nObjective Template List (Approved):");
//Get template description
Console.WriteLine($"\n\nDescription of plan template");
Console.WriteLine($"Number of Fields: {plantemplate.Fields.Count()}");
Console.WriteLine($"Plan Rx: {plantemplate.DosePerFraction} x {plantemplate.FractionCount}");
Console.ReadLine();
}
The result:
Finally, the following code will utilize the ClinicalTemplateReader API to generate an automated plan.
static void Execute(Application app)
{
// Generate plan from template
string server = "Varian Image Server"; // <== update to your image server address e.g. "localhost", "10.11.12.13"
var clinicalProtols = new ClinicalTemplate(server);
var approvedTemplates = clinicalProtols.PlanTemplates.Where(x => x.Preview.ApprovalStatus.Contains("Approved")).ToList();
var plantemplate = approvedTemplates.ElementAt(1); // Select an approved plan template
var _patient = app.OpenPatientById("PatientIDHere"); // Give a valid PatientID1
var _structureSet = _patient.StructureSets.FirstOrDefault(x => x.Id == "ROI");
Console.WriteLine("\n\nGenerating new automation plan");
_patient.BeginModifications();
//Generate plan
var _newplan = clinicalProtols.GeneratePlanFromTemplate(null, _structureSet, plantemplate, null);
//Set Prescription.
clinicalProtols.SetRx(plantemplate, _newplan, "cGy");
string plan_update = $"Plan created with {_newplan.Beams.Count()} beams";
Console.Write($"Generated Plan {_newplan.Id} in {_newplan.Course.Id}\n{plan_update}\nPlan Rx: {_newplan.TotalDose} in {_newplan.NumberOfFractions}fx");
app.SaveModifications();
app.ClosePatient();
Console.ReadLine();
}
To execute this part you must add the following attribute to your code:
[assembly: ESAPIScript(IsWriteable = true)]
The response from the console should be as follows.
After refreshing the patient in Eclipse, you should see a new course, plan, and fields generated on the patient.
A new feature available in Version 1.0.0.3 is the ability to leverage the plan quality metrics in a clinical protocol to evaluate the plan. Below is the code to select a clinical protocol, and based on that clinical protocol determine the pass/fail of each clinical plan. This does not require that you first build the plan with the ClinicalTemplateReader, but rather can be used as a stand-alone feature for getting plan goals. NOTE: This works for both plans and plan sums, so now you can build clinical protocols with goals for an overall plansum and get the objective results!
var reader = new ClinicalTemplate("Varian Image Server");
Console.WriteLine("Please select a clinical protocol:");
int cp_num = 0;
Console.WriteLine("Approval Statuses");
Console.WriteLine($"{String.Join("\n",reader.ClinicalProtocols.Select(x=>x.Preview.ApprovalStatus).Distinct())}");
var clinicalProtocols = reader.ClinicalProtocols.Where(x => x.Preview.ApprovalStatus == "Approved");
foreach (var cps in clinicalProtocols)
{
Console.WriteLine($"[{cp_num}]. {cps.Preview.ID} - {cps.Preview.TreatmentSite}: {cps.Preview.LastModified}");
cp_num++;
}
int cp_selection = Convert.ToInt32(Console.ReadLine());
var clinicalprotocol = clinicalProtocols.ElementAt(cp_selection);
//open patient, course and plan.
var patient = app.OpenPatientById("PatientId");//patient id here
var course = patient.Courses.FirstOrDefault(x => x.Id == "CourseId");//course id here
var plan = course.PlanSetups.FirstOrDefault(x => x.Id == "PlanId");//plan id here
//This is the method that compares the planning item to the clinical protocol.
var metrics = reader.CompareProtocolDoseMetrics(plan, clinicalprotocol);
foreach(var doseMetric in metrics)
{
Console.WriteLine($"{doseMetric.MetricText}\n\tResult: ({doseMetric.Pass}) {doseMetric.ResultText}");
}
The final result should look something like this.
In V16, the Rapidplan details, such as the structures included in a Rapidplan model, can be queried with ESAPI. This feature can now be leveraged by the ClinicalTemplateReader. First set up the code to allow the user to select the Rapidplan model they want to select for a given plan. Here newplan is used as an example as seen above
//optimize plan using Rapidplan.
var rp_models = app.Calculation.GetDvhEstimationModelSummaries();
var rp_count = 0;
foreach(var rp in rp_models)
{
Console.WriteLine($"[{rp_count}]. {rp.Name} - {rp.TreatmentSite}");
rp_count++;
}
Console.WriteLine("Please pick your RP Model:");
int rp_num = Convert.ToInt32(Console.ReadLine());
var rp_model = rp_models.ElementAt(rp_num);
clinicalProtols.OptimizeFromRapidPlanModel(app.Calculation.GetDvhEstimationModelStructures(rp_model.ModelUID), _newplan, rp_model, null, null, false);
if (_newplan.Beams.First().Technique.Id.Contains("ARC"))
{
_newplan.CalculateDose();
}
else
{
_newplan.CalculateLeafMotionsAndDose();
}
app.SaveModifications();
Once the code has run, you should have optimization results driven by the DVH Estimation algorithm.
Some notes regarding this method