CSC-F-027-Training-Data-Template.md

qms_version	2.2.0
sop_id	CSC PR.001
sop_version	2.0.1
template_id	CSC F.027
template_version	2.0.1
record_version
record_id	TD-001
title	Training data

Training Data

General

Template ID	CSC F.027
Template Version	2.0.1
QMS Version	2.2.0
SOP ID	CSC PR.001
SOP Version	2.0.1
Regulatory References

Author
Approval

Purpose

This document describes the data collection and processing procedure for {{device.name}} {{device.version}}, including demographic breakdown of the subjects in training and testing datasets.

Scope

This document applies to data used for the development of {{device.name}}, and includes changes made in release {{device.version}}.

Definitions

Item	Definition

Datasets

1. Data Curation, Verification and Exclusion

[TODO: Identify the sources of the data used in the development of the application. ]

2. Demographics Information and Bias Identification

[TODO: Discuss the demographic splits between sex, ethnicity, age etc and any biases, and how they were handled.]

	Positive	Negative
Male
Female

	Positive	Negative
Not specified
Asian
Black
Latin American
Mixed
White

Male	Positive	Negative
0-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
51-55
56-60
61-65
66-70
71-75
76-80
81-85
86-90
91+

Female	Positive	Negative
0-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
51-55
56-60
61-65
66-70
71-75
76-80
81-85
86-90
91+

3. Anonymisation and Psuedonymisation Methods

[TODO: Outline the psuedonymisation schema, anonymised data, along with the rationale]

4. Dataset Version and Identification

[TODO: State the current version of the dataset used during development ]

5. Data Labelling

[TODO: Describe methods of labelling, the staff responsible, qualification level, any assistive tools used etc]

6. Data Transformation and enrichment

[TODO: Systematic data transformations applied and rationale, eg cropping, normalisation etc ]

7. Data Storage and Access Control

[TODO: who has access and why. Who "gate keeps" access the data]