QUALITY SYSTEM PROCEDURE

Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Design and Development Process for AI Models

1.0 PURPOSE
The purpose of this procedure is to ensure that the organization will plan and control the design and
development of the software is error free and fully compliant to product safety protocols
consistently and efficiently to deliver Best in Class healthcare products that meet and exceed
expectations of customer and meet the quality objectives of the organization.

2.0 SCOPE
Scope of this procedure includes designing of all the products developed at Artificial Learning
Systems India Pvt. Ltd.

3.0 RESPONSIBILITY
Chief Deep Learning Scientist – Responsible for defining and approving the processes and ensuring
the implementation and compliance of the process to meet the quality objectives and regulatory
requirements.

Deep Learning Scientist – Responsible for implementing and reviewing the overall development
process on an ongoing basis

Deep Learning model Developer – Ensure all the design and development processes are followed in
the development of the products

Data Scientist – Ensure all the design and development processes are followed in the testing of the
products

Annotators/Doctors – Responsible for data annotation. To manually classify data into a given
number of classes.

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Process Quality Assurance – To ensure periodic review and audit of design and development
activities and approves design and development process meets the regulatory requirements. To
review and approve any changes to the design and development process.

IT Infrastructure and DevOps – Ensure that all the required infrastructure, tools, software and any
other resources required for the design and development of products are available on time and
meet the regulatory compliance. It also responsible for procuring, maintaining and servicing the
required infrastructure needs.

Support – Responsible for providing post product delivery support and collect, log, analyze and
evaluate the feedback from the customer. Also responsible for periodic survey of the product
deployed in the market.

4.0 PROCEDURE-Software Development Life Cycle

AI Software design and development is dependent on an initial idea, which may come from the
Management, market research, employees, from existing and prospective customer, from industry
requirements or from a previously tried out/developed product.

All the ideas are collected, reviewed and prioritized. The prioritized ideas will be evaluated by key

stakeholders to determine the feasibility of the ideas and required resources to develop the

products. The prioritized ideas will be assigned to the AI design and development team for product

realization

Based on the inputs from AI Design specification product design & development planning is carried
out.

Design & development members are identified & assigned for the activities with the target date.

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Design and development planning process includes plans, reviews, verifications & validations that
are appropriate to the design & development stages.

Stage 1 : Idea generation

PROTOTYPING/ CONCEPT FEASIBILITY PROCESS

 Input :

o Idea and Customer Feedback logs, Market Research, Market Demand

 Process and Plan :

o AI product design and development is dependent on an initial idea, which may come
from the Management, market research, employees, from existing and prospective
customer, from industry requirements or from a previously tried out/developed
product.

o All the ideas are collected, reviewed and prioritized in the Prototyping Process. The

prioritized ideas will be evaluated by key stakeholders to determine the feasibility of the

ideas and required resources to develop the products.

 Output:

o Prioritized ideas that will be assigned to the AI design and development team for

product realization ( Idea Specification document)

Stage 2: Data Collection:

 Data Identification: Input:

o Prioritized Idea

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Process:
o Identify the type of data required for building the product
o Decide the size of data required
Output:
o Type and size of data documented
 Data Gathering
Input:
o Type of data and the size of data
Process:
Identify various data sources (Internal and External)
Data gathering Process: Download data from various open source data base, identify
hospital partners and research partner for data gathering
Store data in central data base with proper controls and configuration
Security and user access defined for the data

Output:

Collected Data stored in the central server data security and access
approval
 Data Cleansing
Input:
o Access to the identified and gathered data
Process
o Data Cleansing activities to remove bad quality data and any other data that
do not qualify for developing the product
o Identify and define the parameters for annotation
o Identify different buckets for annotation and provide necessary access to the
annotators based on the roles and qualification for annotation

Output:

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

o Cleansed data ready for annotation

 Data Annotation
Input:
o Cleansed data with the required access for annotation
o Annotators identified and assigned for data annotation
o Doctors identified for Annotation
Process:
Annotators will be assigned respective buckets
Annotators will annotate each image assigned in their buckets
Annotation by the annotators are peer reviewed
Doctor review sample to be defined and identified
Doctor will review a sample of the annotation from the respective annotators
completed bucket.
Output:
Reviewed and annotated data
 Data Segregation
Input: Reviewed and Annotated data
Process:
Annotated data is divided into Training & Validation set and Test set.
Training and validation set used for training the model.
Training set and validation set should have good coverage of real world data.
Test data set: The quality of the test data set is an important factor which influences
the acceptance of the accuracy metric of the model evaluated. If the test data set in
true sense does not represent the production data, or real world data, or is biased,
then accuracy of the model reported cannot be trusted.
All the models are developed using the standard dataset and results are captured
and saved.
Apart from this we have test set for separate camera makers like Topcon and Forus.
Every-time a new model is developed it is tested against the test dataset.
Output:
o Training and Validation and Test data segregation and frozen for developing Models

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Stage 3: Develop Deep Learning Framework and Models

 Input:
o Software design and development is dependent on an initial inputs from Idea
Specification Document
 Plan and Process:
o Define High level design Architecture and specification model development planning
o Design & development members are identified
o Assigned for the activities with the target date.
o Identify Technology and infrastructure ( process)
o Define product Specification
o Create Models
o Preprocessing for training Model:

o Model Training ( using training data )

o Model verification ( test data)
o Model Retraining
o Fine tuning
o Final base lined Model
Stage 4: Validation Studies:
Validation studies are conducted on the final base lined models as per the Clinical
Evaluation process ALSQSP/27
Final product model integration (Product Design and Development team)
 Output:
o Approved product Model for product integration

Stage 5:Pre-Prod Deployment

 Input
o Pre-Prod Deployment notes to the product team
o Deployment Preparation/Notifications
o Approved Evaluation

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

 Plan and Process

o Prepare Deployment Plan/Notes
o Prepare Rollback Plan/Notes
o Prepare Support Plan
o Notifications:
 Notifying stake holders(Planned deployment-will have a date in prior, Agile
deployment-update the date)

Stage 6: Production Deployment & Support:

 Input
o Deployment Plan for the Model
o Deployment Notification
 Plan and process:
o Monitor Deployment
o Transition to Support (KT)
o Support team to handle production support
o Collect customer feedback and complaints
o Assign tickets from the customer to the AI team to fix production issues.
 Output :
o Handover to Design and Development Product Team

4.3 DESIGN AND DEVELOPMENT of Models REVIEW

 At suitable stages as indicated in the design & development planning, systematic reviews are
conducted to evaluate the ability of design to meet the input requirements & identify any

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

problems for necessary actions by the technical team. These reviews are recorded in the design
& development review record.
4.5 DESIGN AND DEVELOPMENT VERIFICATION
 Verification of design is carried out by QA inspecting & testing for the subsistence of design to
ensure that the design and development outputs have met the design and development input
requirements.
 QA inspection & test reports (D & D verification records) are reviewed & verified by QA
Manager.

4.6 DESIGN AND DEVELOPMENT VALIDATION

 Validation is performed to ensure that the resulting product is capable of meeting the
requirements for the specified application or intended use.
 Records of the results of validation and any necessary actions to be taken are maintained in
design and development validation report.
 Refer Clinical Evaluation process ( ALSQSP/27 ) for Validation of Models

4.7 DESIGN AND DEVELOPMENT TRANSFER

The organization ensures proper transfer of design and development outputs to Delivery team.
Design and development outputs are verified as suitable for delivery before becoming final
specifications and that delivery team capability can meet product requirements.
4.8 CONTROL OF DESIGN AND DEVELOPMENT CHANGES
 Design & development change requested by any department is recorded in the design change
review format and is forwarded to Director for review & approval.
 Design / Document change request can be originated by any department & forwarded to D&D
department (technical) / MR for analysis of design / document change.

Design and Development Process for End Product

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

5.0 PURPOSE
The purpose of this procedure is to ensure that the organization will plan and control the design and
development of the software is error free and fully compliant to product safety protocols
consistently and efficiently to deliver Best in Class healthcare products that meet and exceed
expectations of customer and meet the quality objectives of the organization.

6.0 SCOPE
Scope of this procedure includes designing of all the products developed at Artificial Learning
Systems India Pvt. Ltd.

7.0 RESPONSIBILITY
Design Head – Responsible for defining and approving the processes and ensuring the
implementation and compliance of the process to meet the quality objectives and regulatory
requirements.

Development Head – Responsible for implementing and reviewing the overall development process
on an ongoing basis

Developer – Ensure all the design and development processes are followed in the development of
the products

Tester – Ensure all the design and development processes are followed in the testing of the products

Process Quality Assurance – To ensure periodic review and audit of design and development
activities and approves design and development process meets the regulatory requirements. To
review and approve any changes to the design and development process.

IT Infrastructure and DevOps – Ensure that all the required infrastructure, tools, software and any
other resources required for the design and development of products are available on time and
meet the regulatory compliance. It also responsible for procuring, maintaining and servicing the
required infrastructure needs.

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Support – Responsible for providing post product delivery support and collect, log, analyze and
evaluate the feedback from the customer. Also responsible for periodic survey of the product
deployed in the market.

8.0 PROCEDURE-Software Development Life Cycle

Software design and development is dependent on an initial idea, which may come from the
Management, market research, employees, from existing and prospective customer, from industry
requirements or from a previously tried out/developed product.

All the ideas are collected, reviewed and prioritized in the Prototyping Process. The prioritized ideas

will be evaluated by key stakeholders to determine the feasibility of the ideas and required

resources to develop the products. The prioritized ideas will be assigned to the design and

development team for product realization

Based on the inputs from prototyping process, product design & development planning is carried
out.

Design & development members are identified & assigned for the activities with the target date.

Design and development planning process includes plans, reviews, verifications & validations that
are appropriate to the design & development stages.

Stage 1 : Idea generation

PROTOTYPING/ CONCEPT FEASIBILITY PROCESS

 Input :

o Idea and Customer Feedback logs

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

 Process and Plan :

o Software design and development is dependent on an initial idea, which may come from
the Management, market research, employees, from existing and prospective customer,
from industry requirements or from a previously tried out/developed product.

o All the ideas are collected, reviewed and prioritized in the Prototyping Process. The

prioritized ideas will be evaluated by key stakeholders to determine the feasibility of the

ideas and required resources to develop the products.

 Output:

o Prioritized ideas that will be assigned to the design and development team for product

realization ( Idea Specification document)

Stage 2: High Level Architecture:

 Input:

o Software design and development is dependent on an initial inputs from prototyping

process ( Idea Specification Document)

 Plan and Process:

o Define High level design Architecture and specification for product design & development
planning
o Design & development members are identified
o Assigned for the activities with the target date.
o Identify Technology and infrastructure ( process)
o Define Functional Specification

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

o Create HLA
o Risk Management (process)
o Change control( process)
o Traceability
o Review and Freeze the Architecture
 Output:
o Approved HLA ( output)

Stage 3: Low Level Design

 Input:
o approved HLA is input for LLD

 Plan and Process:

o Define Low level Design and functional design and development planning process
o Assigned for the activities with the target date for Low level design.
o Create LLD
 Define Modules
o Raise request for access to environments (Refer to ALS/QSP/07 Infrustructure)
o Risk Management
o Change control
o Traceability
o Review and Freeze the LLD
 Output
o Approved LLD ( output)

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

Stage 4: Development

 Input
o Approved LLD is the input

 Plan and Process

o Coding for the modules

o Coding integration modules

o Create Unit test plan

o Perform Unit testing

 Fix unit test bugs

 Review Code

 Output:
o Unit tested Coding for the modules
o Unit tested Coding integration modules
o Unit test report ( log)

Stage 5: Testing

 Input
o Approved Idea specifications
o Approved HLA
o Approved LLD (Input)
o Approved Code and UT Results

 Plan and Process

o Create Test Plan

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

o Create Test Scenarios

o Create Test Cases
o Create Integration Test Cases
o Execute System Test and System Integration Test
o Defect Management
 Defect Logging (output)
 Assign Defects/ Prioritize
 Defect Analysis
 Retest / Closure
 Output:
o Test Result Reports(output)
 Defect Analysis report
o Handover to UAT
 UAT Results
 Approved UAT

Stage 6:Pre-Prod Deployment

 Input
o Finalized and tested code[input]
o Pre-Prod Deployment notes
o Deployment Preparation/Notifications
o Approved UAT

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

 Plan and Process

o Prepare Deployment Plan and Release Notes
o Prepare Rollback Plan and Release Notes
o Prepare Support Plan
o Notifications:
 Notifying stake holders(Planned deployment-will have a date in prior, Agile
deployment-update the date)

Stage 7: Production Deployment & Support:

 Input
o Deployment Plan
o Deployment Notification

 Plan and process:

o Monitor Deployment
o Handover Release Notes to support
o Transition to Support (KT)
o Support team to handle production support
o Collect customer feedback and complaints
o Assign tickets from the customer to the development team to fix production issues.

 Output :
o Customer Sign Off

4.3 DESIGN AND DEVELOPMENT REVIEW

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

 At suitable stages as indicated in the design & development planning, systematic reviews are
conducted to evaluate the ability of design to meet the input requirements & identify any
problems for necessary actions by the technical team. These reviews are recorded in the design
& development review record.

4.5 DESIGN AND DEVELOPMENT VERIFICATION

 Verification of design is carried out by QA inspecting & testing for the subsistence of design to
ensure that the design and development outputs have met the design and development input
requirements.
 QA inspection & test reports (D & D verification records) are reviewed & verified by QA
Manager.

4.6 DESIGN AND DEVELOPMENT VALIDATION

 Validation is performed to ensure that the resulting product is capable of meeting the
requirements for the specified application or intended use.
 Records of the results of validation and any necessary actions to be taken are maintained in
design and development validation report.

4.7 DESIGN AND DEVELOPMENT TRANSFER

The organization ensures proper transfer of design and development outputs to Sales team. Design
and development outputs are verified as suitable for delivery before becoming final specifications
and that Sales team capability can meet product requirements. Design Transfer includes training /
demonstration to Sales team on the new product and its key features, transfer of User Manual.

4.8 CONTROL OF DESIGN AND DEVELOPMENT CHANGES

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

 Design & development change requested by any department is recorded in the design change
review format and is forwarded to Director for review & approval.
 Design / Document change request can be originated by any department & forwarded to D&D
department (technical) / MR for analysis of design / document change.

1.0 REFERENCE
ISO 13485: 2016 - CL. 7.3 Design and development

IEC 62304:2006 Relevant sections are mentioned below

Section No. Section Description

4 General Requirements
4.1 Quality Management System
4.2 RISK MANAGEMENT PROCESS complying with ISO 14971
4.3 Software Safety Classification for each software system
5 Software Development PROCESS
5.1 Software Development Planning
5.1.1 Software development plan
5.1.2 Keep software development plan updated
5.1.3 Software development plan reference to SYSTEM design and development
5.1.4 Software development standards, methods and tools planning
5.1.5 Software integration and integration testing planning
5.1.6 Software VERIFICATION planning
5.1.7 Software RISK MANAGEMENT planning
5.1.8 Documentation planning
5.1.9 Software configuration management planning
5.1.10 Supporting items to be controlled

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

5.1.11 Software CONFIGURATION ITEM control before verification

5.2 Software Requirements Analysis
5.2.1 Define and document software requirements from SYSTEM requirements
5.2.2 Software requirements content
5.2.3 Include RISK CONTROL measures in software requirements
5.2.4 Re-EVALUATE MEDICAL DEVICE RISK ANALYSIS
5.2.5 Update SYSTEM requirements
5.2.6 Verify software requirements
5.3 Software ARCHITECTURAL Design
5.3.1 Transform software requirements into an ARCHITECTURE
5.3.2 Develop an ARCHITECTURE for the interfaces of software items
5.3.3 Specify functional and performance requirements of SOUP item
5.3.4 Specify SYSTEM hardware and software required by SOUP item
5.3.5 Identify segregation necessary for RISK CONTROL
5.3.6 Verify software ARCHITECTURE
5.4 Software Detailed Design
5.4.1 Refine SOFTWARE ARCHITECTURE into SOFTWARE UNITS
5.4.2 Develop detailed design for each SOFTWARE UNIT
5.4.3 Develop detailed design for interfaces
5.4.4 Verify detailed design
5.5 Software Unit Implementation and Verification
5.5.1 Implement each SOFTWARE UNIT
5.5.2 Establish SOFTWARE UNIT VERIFICATION PROCESS
5.5.3 SOFTWARE UNIT acceptance criteria
5.5.4 Additional SOFTWARE UNIT acceptance criteria
5.5.5 SOFTWARE UNIT VERIFICATION
5.6 Software Integration and Integration Testing
5.6.1 Integrate SOFTWARE UNITS
5.6.2 Verify software integration
5.6.3 Test integrated software
5.6.4 Integration testing content

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

5.6.5 Verify integration test procedures

5.6.6 Conduct regression tests
5.6.7 Integration test record contents
5.6.8 Use software problem resolution PROCESS
5.7 Software System Testing
5.7.1 Establish tests for software requirements
5.7.2 Use software problem resolution PROCESS
5.7.3 Retest after changes
5.7.4 Verify SOFTWARE SYSTEM testing
5.7.5 SOFTWARE SYSTEM test record contents
5.8 Software Release
5.8.1 Ensure software VERIFICATION is complete
5.8.2 Document known residual anomalies
5.8.3 Evaluate known residual anomalies
5.8.4 Document released VERSIONS
5.8.5 Document how released software was created
5.8.6 Ensure activities and tasks are complete
5.8.7 Archive software
5.8.8 Assure repeatability of software release
6 SOFTWARE MAINTENANCE PROCESS
6.1 Establish Software Maintenance Plan
6.2 Problem and Modification Analysis
6.2.1 Document and EVALUATE feedback
6.2.1.1 Monitor feedback
6.2.1.2 Document and EVALUATE feedback
6.2.1.3 Evaluate PROBLEM REPORT'S affects on SAFETY
6.2.2 Use software problem resolution PROCESS
6.2.3 Analyze CHANGE REQUESTS
6.2.4 CHANGE REQUEST approval
6.2.5 Communicate to users and regulators
6.3 Modification Implementation

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

6.3.1 Use established PROCESS to implement modification

6.3.2 Re-release modified SOFTWARE SYSTEM
7 SOFTWARE RISK MANAGEMENT PROCESS (referenced ISO 14971)
7.1 Analysis of Software Contributing to Hazardous Situations
7.1.1 Identify SOFTWARE ITEMS that could contribute to a hazardous situation
7.1.2 Identify potential causes of contribution to a hazardous situation
7.1.3 Review published SOUP ANOMALY lists
7.1.4 Document potential causes of SOFTWARE ITEM contributing to hazards
7.1.5 Document sequences of events contributing to hazardous situations
7.2 Risk Control Measures
7.2.1 Define RISK CONTROL measures
7.2.2 RISK CONTROL measures implemented in software
7.3 Verification of Risk Control Measures
7.3.1 Verify RISK CONTROL measures
7.3.2 Document any new sequences of events contributing to hazardous situation
7.3.3 Document TRACEABILITY of software HAZARDS
7.4 Risk Management of Software Changes
7.4.1 Analyze changes to MEDICAL DEVICE SOFTWARE with respect to SAFETY
7.4.2 Analyze impact of software changes on existing RISK CONTROL measures
7.4.3 Perform RISK MANAGEMENT ACTIVITIES based on analyses
8 SOFTWARE CONFIGURATION MANAGEMENT PROCESS
8.1 Configuration Identification
8.1.1 Establish means to identify CONFIGURATION ITEMS
8.1.2 Identify SOUP
8.1.3 Identify SYSTEM configuration documentation
8.2 Change Control
8.2.1 Approve CHANGE REQUESTS
8.2.2 Implement changes
8.2.3 Verify changes
8.2.4 Provide means for TRACEABILITY of change
8.3 Configuration Status Accounting

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

9 SOFTWARE PROBLEM RESOLUTION PROCESS

9.1 Prepare Problem Reports
9.2 Investigate the problem
9.3 Advise relevant parties
9.4 Use change control process
9.5 Maintain records
9.6 Analyze problems for trends
9.7 Verify software problem resolution
9.8 Test documentation contents

2.0 TEMPLATES

Idea Logs

Idea prioritization matrix

Idea Specification

High Level Architecture

Low Level Design

Test Plan

Test Scenario Template

Defect log

Defect Analysis

Customer complaint handling

Customer feedback log

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.
 QUALITY SYSTEM PROCEDURE
Document
Effective Date Revision No Page
Number

Page PAGE 21
ALS/QSP/12 12-Dec-19 02 of NUMPAGES
21

TITLE DESIGN AND DEVELOPMENT

3.0 REVISION HISTORY

Revision No. Release Date Reason Prepared By Approved By
R0 1-Apr-19 Initial Implementation Product Manager Senior Management
Revised Design and Business Analyst , Senior Management and
Development procedure Product Team, MR
to capture the essence of Quality Manager
software development for
R1 12-Dec-19
the team to follow the
standard process across
all areas of product
development
Updated Design and Chief Deep Learning Senior Management and
Development process Scientist MR
R2 12-Dec-19
details for AI model
development process

Copyright ©: This document contains confidential information that belongs to ARTIFICIAL LEARNING SYSTEMS INDIA PVT LTD and cannot be
fully or partially copied without the prior written approval of the designated Directors of the organization. The content in this document is intended
only for the internal stakeholders.