Afra Mannan

Academic

Career

Verbatim text

Afra
Cell: 923092615817 |  Email: afra3537@gmail.com
LinkedIn: https://www.linkedin.com/in/afra-mannan
Address: Model Town Street No.6, Faisalabad Road , Okara , Pakistan
PROFESSIONAL PROFILE
I am a motivated and results-oriented graduate with extensive expertise in full-stack web development, WordPress development, and
mobile application development. I have a proven track record of designing scalable, high-performance, and user-friendly digital
solutions across web and mobile platforms. With a strong foundation in AI and machine learning, I actively explore intelligent systems
and data-driven technologies to enhance application functionality, automate processes, and provide actionable insights.
In addition to my technical skills, I bring a keen eye for UI/UX design, creating visually appealing, intuitive, and user-centered
interfaces that elevate the overall user experience. I am adept at working collaboratively in fast-paced environments, translating
complex requirements into innovative solutions, and continuously learning emerging technologies to stay ahead in the ever-evolving
tech landscape. My passion lies in combining creativity with technology to build impactful products that deliver measurable value for
users and businesses alike.
EDUCATION
BS Computer Science
School of Electrical Engineering and Computer Science (SEECS) , Islamabad , 3.7 (2026)
INTERNSHIP EXPERIENCE
National Center of Artificial Intelligence - NCAI Sectt, Pakistan
12-Jun-2023 - 01-Sep-2023
I worked on the project 'Smart Gaze-Based Annotation of Histopathology Images', where we developed a gaze-based annotation
system using eye-tracking technology to capture pathologists’ focus areas on slides. The system processed gaze data, including
fixations and saccades, to automatically generate bounding boxes around regions of interest, significantly improving annotation
efficiency. This approach reduced annotation time by approximately 65% compared to traditional bounding boxes and 85% compared
to pixel-wise annotation, while also providing insights into enhancing labeling quality through advanced gaze-based techniques.
Machine Vision and Intelligent Systems (MachVIS) Lab - SEECS, NUST
16-Jun-2025 - 05-Sep-2025
I contributed to the 'PathoShield' project, which focused on predicting antimicrobial resistance (AMR) from MALDI-TOF mass
spectrometry data. I implemented a deep learning pipeline to build a real-time, uncertainty-aware AI platform for clinical decision
support and outbreak surveillance. The project addressed challenges in delayed diagnostics and non-adaptive surveillance systems
by integrating continually adapting models, thereby improving diagnostic intelligence and enabling timely, evidence-based clinical
decisions.
FINAL YEAR PROJECT
PathoShield: AI-Driven Antimicrobial Resistance Prediction from MALDI-TOF Data
PathoShield is a deep learning-based platform designed to predict antimicrobial resistance (AMR) in real-time using MALDI-TOF
mass spectrometry data. Addressing the growing global threat of AMR, the system overcomes challenges posed by delayed
diagnostics, inappropriate antibiotic use, and fragmented, non-adaptive surveillance systems. By leveraging uncertainty-aware AI
models, PathoShield provides accurate, interpretable predictions that support timely clinical decision-making and enable healthcare
professionals to prescribe effective treatments with confidence. The platform also incorporates adaptive learning to continuously
update its predictive models, facilitating real-time outbreak monitoring and enhancing public health responses. Through this
integrated approach, PathoShield not only improves diagnostic efficiency but also empowers clinicians and healthcare systems to
proactively tackle resistant infections, bridging the gap between laboratory data and actionable clinical insights.

AI enrichment