Mujibur Rahman Chowdhury, PhD

www.linkedin.com/in/mujibur-rahman-chowdhury-ph-d-62932247

Email: ad0we7@r.postjobfree.com, Phone: 409-***-****.

Profile Summary

• Boasting over 6 years of specialized, hands-on experience in Data Science, Machine Learning, and Deep Learning, I am keen on mastering the complete Machine Learning workflow and deepening my expertise in MLOps and CI/CD pipelines. I’m also enthusiastic about learning and mastering emerging technologies like Generative AI such as NLP, Large Language Models (LLM), and GANs.

• Possess comprehensive expertise in structuring, analyzing, and leveraging data in both structured and unstructured environments to drive a variety of insights/algorithms. I’m skilled in employing statistical modeling techniques such as decision trees and logistic regression to enhance product/system performance, quality, and accuracy.

• I hold in-depth knowledge of image processing, including techniques like denoising, deblurring, and deconvolution. Proficient with a range of classification, detection, and segmentation models like VGGNet, MobileNet, ResNet, Faster RCNN, Mask RCNN, and Yolo instance.

• Proficient in translating complex algorithms and technical specifications into efficient and readable code, using languages and tools like Python, MATLAB, OpenCV, Scikit-learn, TensorFlow, Keras, Pandas, Matplotlib, Seaborn, Hugging Face Transformers, and NLTK.

• Experienced with various cloud platforms and services, including AWS SageMaker, S3, and Amazon Rekognition, and well-versed in large-scale computing frameworks, data analysis systems, and modeling environments.

• A dedicated and committed team player with a proven history of effective collaboration in cross-functional environments.

Professional Experience

Bayer AG - Crop Science St. Louis, MO

Computer Vision Scientist (Machine Learning & Imaging Data Scientist) April 2022 - Present

• Designed an ML model for cotton seed classification, achieving four levels of damage identification and boosting North American revenue by $18 million within a year.

• Devised an ML model to detect internal damage in cotton seeds, achieving 98% accuracy.

• Engineered an algorithm for corn seed crack/damage detection, effectively identifying both internal and external imperfections via x-ray imagery.

• Enhanced model performance through image pre-processing techniques such as denoising and deblurring.

• Developed a segmentation method using the watershed algorithm for isolating individual seeds in bulk DICOM images, thus enabling more accurate ML analyses.

• Contributed to agriculture sustainability by devising an ML model to classify rice quality, helping farmers cultivate higher germination rates and better yields.

Oregon State University Corvallis, OR

Research Scientist/Visiting Assistant Professor November 2021 - August 2022

• Pioneered the application of advanced algorithms for high-fidelity image reconstruction in CT and MRI scans, significantly improving diagnostic accuracy.

• Developed and optimized cutting-edge Image Super-Resolution techniques, transforming low-res medical images into high-res versions to enable more precise medical assessments. University of Texas at Dallas Richardson, TX

Research Associate/ Data Scientist August 2020 – November 2021

• Elevated to a more senior role for exceptional performance in project acceleration and high-quality research publications in industry-academia recognized journals.

• Engineered state-of-the-art algorithms for medical image reconstruction in CT and MRI, achieving breakthroughs in image quality under challenging conditions such as Poisson noise.

• Pioneered L1/L2 regularization techniques for image super-resolution and deconvolution, contributing to more accurate and detailed medical diagnoses.

University of Texas at Dallas Richardson, TX

Research Assistant August 2015 - August 2020

• Devised and optimized noise and blur reduction algorithms for photon-count images, leading to a significant improvement in PSNR metrics (0.5 dB to 2.0 dB), thereby enhancing the overall image quality for better diagnostic outcomes.

UT Southwestern Medical Center Dallas, TX

Student Intern June 2016 - August 2016

• Developed an innovative system matrix for PET (Positron Emission Tomography) Image Reconstruction, achieving a near-perfect 91% accuracy rate. This contribution significantly enhanced the diagnostic capabilities of PET imaging systems.

Education

Ph.D. in Applied Mathematics August 2020

University of Texas at Dallas, Richardson, TX Specialization: Image Processing M.S. in Data Science December 2019

University of Texas at Dallas, Richardson, TX Concentration: Mathematics for Advanced Analytics Graduate Certificate in Data Science August 2018

University of Texas at Dallas, Richardson, TX Focusing on: Big Data Analytics, Machine Learning Algorithms Technical Skills

Languages: Proficient in Python, MATLAB; Familiar with R, SQL. Machine Learning: Experienced in Regression, SVM, PCA, K-means, CNN; Knowledgeable in Optimization Techniques including Gradient Descent and Cross Validation. Professional Development: Completed advanced courses in Python, Deep Learning, SQL Programming, and Convex Optimization from platforms like Udemy, Coursera, and LinkedIn Learning. Outreach Service and Leadership

• Organizer, Workshop on Recent Developments on Mathematical/Statistical approaches in Data Science

(MSDAS), The University of Texas at Dallas, June 2019.

• President, SIAM UT Dallas and SMU Student Chapter, SIAM stands for Society for Industrial and Applied Mathematics, the largest professional organization for applied mathematics, 2017-2019.

• Founder and President, Society for Awareness of Youth (SAY), 2011-2013. Some Publications

1. M. R. Chowdhury, J. Qin and Y. Lou. Non-blind and Blind Deconvolution under Poisson Noise using Fractional-order Total Variation. Journal of Mathematical Imaging and Vision (JMIV) 62, pages 1238–1255 (2020).

2. M. R. Chowdhury, J. Zhang, J. Qin and Y. Lou. Poisson Image Denoising Based on Fractional-Order Total Variation. Inverse Problem and Imaging February 2020, 14(1): 77-96. 3. E. Islam and M. R. Chowdhury. Persishable Inventory System at service facility with N-policy consider reneging and rejection of Pool Customers. Proceedings of the International Conference on Applied Mathematical Models ICAMM 2014; page : 117-126, January 3-5, India. 4. M. R. Chowdhury and K. Das. Application of Singular Value decomposition and non-negative matrix factorization in the image compression. MS thesis, Lamar University, 2015. Grants and Award

• PhD Research Small Grants, 2020. • SIAM Student Chapter Certificate of Recognition 2020.

• SIAM Student Travel Award. SIAM Conference on Imaging Science, Canada. • Dean’s Award 2009, CU.

Contact this candidate