Onco-Phenotype inference information

Azure AI Health Insights Onco-Phenotype model was trained with labels that conform to the following standards.

• Tumor site and histology inferences: WHO ICD-O-3 representation.
• Clinical and pathologic stage TNM category inferences: American Joint Committee on Cancer (AJCC)'s 7th edition of the cancer staging manual.

You can find an overview of the response values here:

Inference type	Description	Values
tumorSite	The tumor site	None, ICD-O-3 tumor site code (e.g. C34.2)
histology	The histology code	None, 4-digit ICD-O-3 histology code
clinicalStageT	The T category of the clinical stage	None, T0, Tis, T1, T2, T3, T4
clinicalStageN	The N category of the clinical stage	None, N0, N+
clinicalStageM	The M category of the clinical stage	None, M0, M1
pathologicStageT	The T category of the pathologic stage	None, T0, Tis, T1, T2, T3, T4
pathologicStageN	The N category of the pathologic stage	None, N0, N+
pathologicStageM	The M category of the pathologic stage	None, M0, M1

Confidence score

Each inference has an attribute called confidenceScore that expresses the confidence level for the inference value, ranging from 0 to 1. The higher the confidence score is, the more certain the model was about the inference value provided. The inference values should not be consumed without human review, no matter how high the confidence score is.

Importance

When you set the includeEvidence property to true, each evidence property has an importance attribute that expresses how important that evidence was to predicting the inference value, ranging from 0 to 1. A higher importance value indicates that the model relied more on that specific evidence.

Next steps

To get better insights into the request and responses, read more on following page:

Model configuration