Go to AIH-HSP AI Modules

Function
Ref. Model
I/O Data
SubAIMs
JSON MData
Profiles
Ref. Software
Conformance
Performance

1 Functions

The Neural Network Model (AIH‑NNM) AIM wraps a Neural Network Model and uses it to process AIH Data: it gives the AIH Data resulting from the processing of the AIH Data it receives. The Neural Network Model learns as the AIM processes AIH Data. The AIM is what is exchanged: the Health Back End makes the AIM that wraps its model available to the Health Front Ends; a Health Front End returns it with the Neural Network Model trained, and the AIM gives the Neural Network Model it holds for that; the Health Back End uses the models it receives for federated learning. The AIM:

Receives AIH Data Data to be processed by the model.
Receives ML Model Object The Neural Network Model to be wrapped or updated.
Produces AIH Data The AIH Data resulting from the processing of the AIH Data received.
Produces ML Model Object The Neural Network Model the AIM holds, trained on the AIH Data it processed.

2 Reference Model

Figure 1 depicts the Reference Model of the Neural Network Model (AIH‑NNM) AIM.

Neural Network Model AIH-NNM AIM

Figure 1 – Reference Model of the Neural Network Model (AIH‑NNM) AIM

3 I/O Data

Table 1 specifies the Input and Output Data of the Neural Network Model (AIH‑NNM) AIM.

Table 1 – I/O Data of the Neural Network Model (AIH‑NNM) AIM

Input Description
AIH Data Data to be processed by the model.
ML Model Object The Neural Network Model to be wrapped or updated.
Output Description
AIH Data The AIH Data resulting from the processing of the AIH Data received.
ML Model Object The Neural Network Model the AIM holds, trained on the AIH Data it processed.

4 SubAIMs

No SubAIMs.

5 JSON Metadata

https://schemas.mpai.community/AIH1/V1.1/AIMs/NeuralNetworkModel.json

6 Profiles

No Profiles.

7 Reference Software

Under development.

8 Conformance Testing

Table 2 provides the Conformance Testing Method for the Neural Network Model (AIH‑NNM) AIM.

If a schema contains references to other schemas, conformance of data for the primary schema implies that any data referencing a secondary schema shall also validate against the relevant schema, if present, and conform with the Qualifier, if present.

Table 2 – Conformance Testing Method for the Neural Network Model (AIH‑NNM) AIM

Receives AIH Data Shall validate against AIH Data schema.
Receives ML Model Object Shall validate against ML Model Object schema.
Produces AIH Data Shall validate against AIH Data schema.
Produces ML Model Object Shall validate against ML Model Object schema.

9 Performance Assessment

Not part of this specification.

Go to AIH-HSP AI Modules