Is my use case diagram about AI expertizer interpretable?

Question

I have this Use-case diagram:

Stop Scroll Here!!!

What is your first impression about the use case? I mean is it interpretable intuitively? After you answer it, I will explain what that use case is talking about.

Explanation (Spoiler Alert)

Based on SRS description:

The Fetal Expertizer is an advanced system designed to assist healthcare professionals in detecting fetal abnormalities through image analysis, primarily using ultrasound data. Powered by deep learning algorithms, it provides real-time insights into potential issues, improving diagnostic accuracy and ultimately enhancing patient outcomes.

The system will be built on Amazon Web Services (AWS) to leverage its scalability, reliability, and security. Key services in the tech stack will include AWS Fargate for containerized microservices, ensuring the system is highly available and can scale efficiently. For backend data storage, the system will use Amazon DynamoDB, a fully managed NoSQL database, to store common application data, enabling fast and flexible access to data across different parts of the system. Other services include Amazon S3 for data storage, Amazon SageMaker for model training and evaluation, and AWS Lambda for serverless computing. These services will work together to ensure smooth performance, scalability, and flexibility as the system evolves.

A distinctive feature of the Fetal Expertizer is its role-based operational model, which defines clear responsibilities for various actors: Acquisitors collect the data, Annotators label the data, Validators verify both collected and labeled data, Trainers improve the models, Evaluators assess model performance, and Administrators manage system configurations and user roles. The Model Consumer role is flexible, encompassing both healthcare professionals who use the system for diagnostic insights and patients who can access their own reports. This approach ensures that the system is tailored to the needs of various stakeholders, driving continuous improvement and making it adaptable to real-world clinical workflows.

Question

Is my use-case correct? Mainly about how I use extend and include keyword. Additionaly about its interpretability.

Answer 1

Sorry if this is not the answer you expect, but no, this use-case is incorrect:

• It misuses «include» to emulate sequences of steps. This is not the purpose of a use case diagram, and results in strange interpretations , such as, by transitive closure of the inclusion, that Upload image for training purpose would include Train model and even the use of the deployed model by its consumers. This seems misleading, as use of the deployed model is independent of the upload of training material.

• It may misuse «extend» to emulate user interface design or again workflow dependencies. But it could also be a legitimate use.

I recommend to revisit the diagram, simplifying it by focusing only on higher level goals, e.g. train model, and analyse image with the model.

In a second step consider activity diagrams to model the sequence of actions for a use case, e.g. for training the model: upload-> labelling-> (optional quality assessment -> training -> deployment

If you want some more details in the Use case diagram, you could be tempted to use functional decomposition, by showing that training includes capturing images, labelling them etc. But functional decomposition is not recommended, as it leads to complex diagrams that nobody understands anymore except the modeler.

Answer 2

My first impression is that you don't really understand what a use-case diagram shows.

A use-case diagram doesn't show how a single use-case works. It can best be regarded as a graphical table-of-contents for the use-cases that exist within the system and the principle actors that interact with the system.