Azka Rehman | AI Research Engineer

Hi, I'm Azka Rehman.

A

Self-driven, quick starter, passionate programmer with a curious mind who enjoys solving a complex and challenging real-world problems.

About

I am an Electrical Engineer Graduated from National University of Science and Technology (NUST). I enjoy problem-solving and coding. Always strive to bring 100% to the work I do. I have worked on technologies like Python, Flask, MySQL, MongoDB, I am passionate about developing complex applications that solve real-world problems impacting millions of users.

Languages: Python, C++, ReactJs
Databases: MySQL, MongoDB
Libraries: NumPy, Pandas, OpenCV, Scikit Learn, Scikit Image, Vtk, SimpleITK
Frameworks: Flask, Keras, TensorFlow, PyTorch

Looking for a fully funded MS opportunity to work in a challenging position combining my skills in AI and Computer Science Field, which provides professional development, interesting experiences and personal growth.

Publications

Healthhub

AI Research Engineer

Research, experiment, and implement the state-of-the-art (SOTA) ML/DL algorithms in the field of medical image analysis.
Development of computer vision and deep learning based medical diagnosis solutions.
Responsible for integrating the developed AI solutions with the Healthhub DICOMLINK and improving the pipeline
Responsible for AI Certification by publishing techniques for different medical imaging products
• Worked with CT, US, and XRAY modalities with different body parts
Tools: Python, Flask, MongoDB, ReactJs, Keras, Pytorch

January 2021 - Present | Seoul, South Korea

Experience

Healthhub

AI Research Engineer

Research, experiment, and implement the state-of-the-art (SOTA) ML/DL algorithms in the field of medical image analysis.
Development of computer vision and deep learning based medical diagnosis solutions.
Responsible for integrating the developed AI solutions with the Healthhub DICOMLINK and improving the pipeline
Responsible for AI Certification by publishing techniques for different medical imaging products
• Worked with CT, US, and XRAY modalities with different body parts
Tools: Python, Flask, MongoDB, ReactJs, Keras, Pytorch

January 2021 - Present | Seoul, South Korea

Projects

Mandibular Canal Segmentation

Automated CAD system for mandibular canal segmentation.

Accomplishments

Tools: Python, Keras, Image processing/li>
Register/login to the web app(with OAuth-based Google Sign-In).
Search and filter songs based on language and singer.
Create multiple playlists and add/remove songs to/from playlist.
Scroll through recently played/viewed songs.

Quiz Web-App

The goal of the project is extraction of medicine names in patient discharge summaries as part of the Harvard Medical School n2c2 challenge

Accomplishments

Skills: NLP, Deep Learning, Text Processing, Medical Entities Recognition
Recognizing Named Entities in medical records is not same as other Name Entities recognition tasks because medical records have their own format of storing information about patient’s disease, and previous diagnosis. Wording used in medical records is a lot different from other text documents and many abbreviations such as CXR, PA, mg and alot of chemical notations are used which are difficult to understand even for non professionals. The publicly available data for medical NER is quite low as compared to other NER datasets. Since data is low we cannot use any complex network because it would overfit and we will not be able to make relaible inference. We cannot use transfer learning because there is a lot of difference between medical records and other text documents. So we have to design a network by carefully selecting it’s parameter so that network learns a generalized pattern from training data.