PCS254- Human Centred Computing Lab
Early Lung Cancer Prediction
PCS254- Human Centred Computing Project Report
Mid-Semester Evaluation
Submitted by:
8024320077 Ram Ray Singh
8024320085 Riya Chaudhary
8024320087 Rohit Singla
8024320091 Samriddhi Kathayat
ME(CSE) First Year
Team Id: C
Submitted to:
Name of the Faculty: Dr Harkiran Kaur
Computer Science and Engineering Department
TIET, Patiala
March 2025
S.No. Assignment Page No.
1 Requirements
Specification Phase
1.1 Stakeholder requirements
and prototypes for your
applications (TDD)
1.2 User stories for all the
required functions in the
system
2 Architectural Design
2.1 Module Structure Chart
2.2 Screen Design Layout of
GUI
2.3 Navigational Design
(Hierarchical Diagram /
Network Diagram)
3 Detailed Design
3.1 Use Case Diagram
3.2 Design Rationale –
GIBIS and QOC
representation
3.3 GUI Design
(Screenshots)
1. Requirement Specification Phase
1.1 Stakeholder requirements and prototypes for your applications
(TDD)
Early Prediction and Prevention of Lung Cancer Using AI
The project "Prevention and Early Detection of Lung Cancer Using AI" focuses on using
advanced artificial intelligence techniques to enhance early detection and prevention of lung
cancer, thereby improving patient outcomes. The system employs machine learning algorithms
trained on diverse medical datasets, including imaging scans, clinical records, and genetic
markers, to identify potential cancerous patterns at an early stage. By integrating predictive
analytics with risk assessment models, the project aims to support healthcare professionals in
making accurate, timely diagnoses and formulating personalized prevention strategies. This AI-
powered approach holds the potential to significantly reduce lung cancer mortality rates through
proactive monitoring, early intervention, and efficient resource utilization.
Need for the Project:
Lung cancer is one of the deadliest forms of cancer, mostly because it’s usually discovered too
late. By the time symptoms show up, the disease has often spread, leaving patients with fewer
treatment options and lower chances of survival. This makes early detection absolutely critical,
yet traditional methods often miss the warning signs until it’s too late.
This is where Artificial Intelligence (AI) comes in. AI has the ability to analyze medical scans,
patient histories, and risk factors with incredible accuracy and speed. It can spot tiny changes in
lung scans that might be invisible to the human eye, or assess someone’s risk based on their
lifestyle, environment, or family history. By doing this, AI can help doctors catch lung cancer
early, when it’s far more treatable, and even work toward preventing it in high-risk individuals.
This project matters because it can genuinely save lives. It’s not just about technology; it’s about
giving people a fighting chance by catching the disease early or stopping it before it even starts.
It also reduces the stress and financial burden that late-stage cancer brings to patients and their
families. With AI, we have the tools to take a smarter, more proactive approach to tackling lung
cancer—and that could change everything.
Collect medical scans (CT, X-rays) and patient details (age, smoking history, family
background).
Functional Requirements
1. Data Collection and Integration
● Collect patient data, including medical history, demographics, and lifestyle factors.
● Integrate with diagnostic systems like imaging devices (X-rays, CT scans).
● Import lab test results, pathology reports, and genetic data if available.
2. AI Model Development and Training
● Train AI models using medical imaging datasets and clinical data.
● Ensure the AI model can identify lung cancer patterns and risk factors.
● Implement continuous model retraining based on new patient data.
3. Quantum ML
In your Early Prediction and Prevention of Lung Cancer Using AI project, Quantum Machine
Learning (Quantum ML) can provide enhanced computational efficiency and accuracy for
complex medical predictions and image analysis.
4. Image Analysis
● Perform automated analysis of CT scans and X-ray images.
● Detect early signs of lung cancer, such as nodules and abnormal tissue patterns.
● Highlight areas of concern for radiologists.
5. User Interface
● Provide a user-friendly interface for healthcare professionals to access AI insights.
● Include a mobile-friendly portal for patients to view recommendations and reminders.
6. Sign up and Login
Allow new users to create an account by providing necessary information (e.g., name, email,
password). Authenticate registered users by validating their email and password.
Maintain secure session management with automatic timeouts.
Provide features for password recovery in case of forgotten credentials.
Non-Functional Requirements
1. Scalability
● Support increased data storage as patient records and image datasets grow.
● Handle a growing number of concurrent users as hospitals and clinics adopt the system.
2. Reliability
● Ensure accurate and consistent results across different datasets and user interactions.
● Include mechanisms for detecting and correcting errors during predictions.
3. Maintainability
● The system should allow for seamless updates without downtime.
● Maintain clear documentation for AI model retraining and software maintenance.
4. Compliance Requirements
● Adhere to medical device regulations for AI-based diagnostic tools.
● Ensure compliance with regional and international healthcare data privacy laws.
5. Logging and Monitoring
● Log all user interactions and prediction outputs for audit purposes.
● Monitor system performance and alert administrators for anomalies.
1.2 User stories for all the required functions in the system
User Cards:
1. User Story: Patient Registration
Card:
As a non registered user ,I want to register myself ,so I can create an account in the system.
Conversation:
The system should collect basic details such as name, age, medical history, and contact
information.
Provide validation for mandatory fields and secure password requirements.
Confirmation:
Successful account creation redirects to the dashboard.
Verify that the database stores the information securely.
Ensure error messages display for missing or invalid fields.
2. User Story: Login Access
Card:
As a registered patient, I want to log in so I can view my health records and prediction
reports.
Conversation:
Support email and password-based login with session management.
Provide password recovery functionality.
Confirmation:
Successful login takes the user to their dashboard.
Invalid credentials trigger appropriate error messages.
Test password reset functionality.
3. User Story: AI-Based Risk Prediction
Card:
As a healthcare professional, I want to upload lung CT scans so the system can predict the
likelihood of lung cancer.
Conversation:
Support image upload in DICOM format.
The AI model should analyze the image and provide a risk prediction with confidence scores.
Confirmation:
The system displays prediction results within 3 seconds.
Confirm accuracy by testing the AI model on a sample dataset.
Validate image processing without errors.
4. User Story: Risk Assessment Report
Card:
As a patient, I want to receive a detailed assessment report so I can understand my lung
cancer risk and prevention recommendations.
Conversation:
The report should include risk scores, detected abnormalities, and prevention tips.
Provide a printable and downloadable version of the report.
Confirmation:
Verify that all report sections are populated correctly.
Test for clear, comprehensible language and actionable recommendations.
Confirm the download functionality works properly.
5. User Story: Data Privacy and Security
Card:
As an administrator, I want to ensure data is encrypted so patient information remains secure.
Conversation:
Encrypt all sensitive information during storage and transmission.
Implement role-based access control to protect patient data.
Confirmation:
Test encryption mechanisms for stored and transmitted data.
Confirm restricted access for unauthorized users.
2. Architectural Design
2.1 Module Structure Chart
2.2 Screen Design Layout of GUI
2.3 Navigational Design (Hierarchical Diagram / Network Diagram)
3. Detailed Design
3.1 Use Case Diagram
Fig- Use case diagram of Early Lung Cancer Prediction
3.2 Design Rationale – GIBIS and QOC representation
Fig - Qoc Diagram of Early Lung Cancer Prediction
3.3 GUI Design (Screenshots):
Fig 3a: Login Page
Fig 3b: Registration Page
Fig 3c: Home Page
