staging-algorithm-eod-public

This project has been moved to staging-client-java.

Extent of Disease (EOD) is a set of three data items that describe how far a cancer has spread at the time of diagnosis. EOD 2018 is effective for cases diagnosed in 2018 and later.

In each EOD schema, valid values, definitions, and registrar notes are provided for

EOD Primary Tumor
EOD Lymph Nodes
EOD Mets
Summary Stage 2018
Site-Specific Data Items (SSDIs), including grade, pertinent to the schema

For cancer cases diagnosed January 1, 2018 and later, the NCI SEER program will collect Extent of Disease (EOD) revised for 2018 and Summary Stage 2018. The schemas have been developed to be compatible with the AJCC 8th Edition chapter definitions.

All of the standard setting organizations will collect the predictive and prognostic factors through Site Specific Data Items (SSDIs). Unlike the SSFs, these data items have formats and code structures specific to the data item.

Download

Java 8 is the minimum version required to use the library.

Download the latest JAR or grab via Maven:

<dependency>
    <groupId>com.imsweb</groupId>
    <artifactId>staging-algorithm-eod-public</artifactId>
    <version>2.1.2</version>
</dependency>

or via Gradle:

compile 'com.imsweb.com:staging-algorithm-eod-public:2.1.2'

Usage

Full documentation can be found in the Wiki

Get a `Staging` instance

Everything starts with getting an instance of the Staging object. There are DataProvider objects for each staging algorithm and version. The Staging object is thread safe and cached so subsequent calls to Staging.getInstance() will return the same object.

For example, to get an instance of the EOD algorithm

Staging staging = Staging.getInstance(EodDataProvider.getInstance(EodVersion.LATEST));

Schemas

Schemas represent sets of specific staging instructions. Determining the schema to use for staging is based on primary site, histology and sometimes additional discrimator values. Schemas include the following information:

schema identifier (i.e. "prostate")
algorithm identifier (i.e. "eod_public")
algorithm version (i.e. "1.4")
name
title, subtitle, description and notes
schema selection criteria
input definitions describing the data needed for staging
list of table identifiers involved in the schema
a list of initial output values set at the start of staging
a list of mappings which represent the logic used to calculate staging output

To get a list of all schema identifiers,

Set<String> schemaIds = staging.getSchemaIds();

To get a single schema by identifer,

Schema schema = staging.getSchema("prostate");

Tables

Tables represent the building blocks of the staging instructions specified in schemas. Tables are used to define schema selection criteria, input validation and staging logic. Tables include the following information:

table identifier (i.e. "ajcc7_stage")
algorithm identifier (i.e. "eod_public")
algorithm version (i.e. "1.4")
name
title, subtitle, description, notes and footnotes
list of column definitions
list of table data

To get a list of all table identifiers,

Set<String> tableIds = staging.getTableIds();

That list will be quite large. To get a list of table indentifiers involved in a particular schema,

Set<String> tableIds = staging.getInvolvedTables("prostate");

To get a single table by identifer,

Table table = staging.getTable("ajcc7_stage");

Lookup a schema

A common operation is to look up a schema based on primary site, histology and optionally one or more discriminators. Each staging algorithm has a SchemaLookup object customized for the specific inputs needed to lookup a schema.

Here is a lookup based on site and histology.

// test valid combinations that do not require a discriminator
EodSchemaLookup lookup = staging.lookupSchema(new EodSchemaLookup("C629", "9231"));
assertEquals(1, lookup.size());
assertEquals("soft_tissue_other", lookup.get(0).getId());

If the call returns a single result, then it was successful. If it returns more than one result, then it needs a discriminator. Information about the required discriminator is included in the list of results. In the followung example, the field discriminator_1 and discriminator_2 are used as a discriminators.

lookup = staging.lookupSchema(new EodSchemaLookup("C111", "8200"));
assertEquals(3, lookup.size());
assertEquals(new HashSet<>(Arrays.asList("oropharynx_hpv_mediated_p16_pos", "nasopharynx", "oropharynx_p16_neg")),
        lookup.stream().map(StagingSchema::getId).collect(Collectors.toSet()));
assertEquals(new HashSet<>(Arrays.asList("discriminator_1", "discriminator_2")), 
        lookup.stream().flatMap(d -> d.getSchemaDiscriminators().stream()).collect(Collectors.toSet()));

// test valid combination that requires discriminator and a good discriminator is supplied
schemaLookup = new EodSchemaLookup("C111", "8200");
schemaLookup.setInput(EodInput.DISCRIMINATOR_1.toString(), "1");
lookup = staging.lookupSchema(schemaLookup);
assertEquals(1, lookup.size());
assertEquals(new HashSet<>(Collections.singletonList("discriminator_1")), 
        lookup.stream().flatMap(d -> d.getSchemaDiscriminators().stream()).collect(Collectors.toSet()));
assertEquals("nasopharynx", lookup.get(0).getId());

schemaLookup.setInput(EodInput.DISCRIMINATOR_1.toString(), "2");
schemaLookup.setInput(EodInput.DISCRIMINATOR_2.toString(), "1");
lookup = staging.lookupSchema(schemaLookup);
assertEquals(1, lookup.size());
assertEquals(new HashSet<>(Arrays.asList("discriminator_1", "discriminator_2")), 
        lookup.stream().flatMap(d -> d.getSchemaDiscriminators().stream()).collect(Collectors.toSet()));
assertEquals("oropharynx_p16_neg", lookup.get(0).getId());

Calculate stage

Staging a case requires first knowing which schema you are working with. Once you have the schema, you can tell which fields (keys) need to be collected and supplied to the stage method call.

An EodStagingData object is used to make staging calls. All inputs to staging should be set on the EodStagingData object and the staging call will add the results. The results include:

output - all output values resulting from the calculation
errors - a list of errors and their descriptions
path - an ordered list of the tables that were used in the calculation

EodStagingData data = new EodStagingInputBuilder()
		.withInput(EodInput.PRIMARY_SITE, "C250")
		.withInput(EodInput.HISTOLOGY, "8154")
		.withInput(EodInput.DX_YEAR, "2018")
		.withInput(EodInput.TUMOR_SIZE_SUMMARY, "004")
		.withInput(EodInput.NODES_POS, "03")
		.withInput(EodInput.EOD_PRIMARY_TUMOR, "500")
		.withInput(EodInput.EOD_REGIONAL_NODES, "300")
		.withInput(EodInput.EOD_METS, "10").build();

// perform the staging
staging.stage(data);

assertEquals(StagingData.Result.STAGED, data.getResult());
assertEquals("pancreas", data.getSchemaId());
assertEquals(0, data.getErrors().size());
assertEquals(12, data.getPath().size());
assertEquals(8, data.getOutput().size());

// check outputs
assertEquals(EodDataProvider.EodVersion.LATEST.getVersion(), data.getOutput(EodOutput.DERIVED_VERSION));
assertEquals("7", data.getOutput(EodOutput.SS_2018_DERIVED));
assertEquals("00280", data.getOutput(EodOutput.NAACCR_SCHEMA_ID));
assertEquals("4", data.getOutput(EodOutput.EOD_2018_STAGE_GROUP));
assertEquals("T1", data.getOutput(EodOutput.EOD_2018_T));
assertEquals("N1", data.getOutput(EodOutput.EOD_2018_N));
assertEquals("M1", data.getOutput(EodOutput.EOD_2018_M));
assertEquals("28", data.getOutput(EodOutput.AJCC_ID));

About SEER

The Surveillance, Epidemiology and End Results (SEER) Program is a premier source for cancer statistics in the United States. The SEER Program collects information on incidence, prevalence and survival from specific geographic areas representing 28 percent of the US population and reports on all these data plus cancer mortality data for the entire country.

License

imsweb/staging-algorithm-eod-public

ErrorLooks like something went wrong!

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases 25

Packages 0

Uh oh!

Languages

Error
Looks like something went wrong!

Packages