Data Science Biomedical Engineering

Khushi Shah

Atlanta, GA 404-***-**** ********@******.*** LinkedIn

EDUCATION

Georgia Institute of Technology, Atlanta, GA GPA:3.75/4 Master of Science - Biomedical Engineering Aug 2024 – present Coursework: 3D printing, Statistical Methods, MRI, Systems Physiology, ML in Biosciences, Human Neuroanatomy. University of Mumbai, Mumbai, Maharashtra GPA: 9.51/10 Bachelor of Engineering - Biomedical Engineering (honors- Data Science) June 2020 - May 2024 Coursework: Biomedical Instrumentation, Medical Sensors, Biomedical Device Regulation, Microcontrollers & embedded systems, Digital Signal Processing, Medical Imaging, Digital Image Processing, Integrated Circuit Design.. SKILLS

Technical: 3D Bioprinting (DLP), CAD (TinkerCAD, SolidWorks), Python, R, MATLAB, Streamlit, TensorFlow, PyTorch, AutoCAD Testing & Analysis:

Mechanical Testing (DMA, compression), Microstructure Imaging (SEM, Micro-CT), AI/ML for Design, CT/MRI Segmentation. Medical AI Testing: SSIM validation, DICOM preprocessing, Siemens MRI data pipelines, and regulatory compliance (ISO 13485, 21 CFR 820).

Regulatory & Quality: Design Control, ISO 14971, FMEA, V&V, Risk Management, Traceability Matrix Soft Skills: Collaboration, Clinical Observation, Technical Communication, Problem Solving Software: MATLAB-Simulink, LTspice, AutoCAD, MS Office, Excel, Word, PowerPoint, C, Python, R Libraries & Tools: TensorFlow, PyTorch, Scikit-learn, Pandas, Numpy, OpenCV, Hover-Net, pydicom WORK EXPERIENCE

Radiology Startup Atlanta, GA

Research Intern Jan 2025 - present

Validated a python-based implementation of MIDAS (Medical Image Data Analysis Suite) modules (EPSI/LITE) by comparing outputs against legacy IDL-dependent MIDAS using SSIM (Structural Similarity Index). Evaluated performance on brain 500+ sMRI scans (T1/T2-weighted) from 3 hospitals (UoM, Emory, JHU) acquired via Siemens 3T/7T MRI machines, focusing on SI/SI_Ref image pairs for quantitative validation. Shadowed MRI scans for patients. Conducted end to end testing across CLI and GUI versions to ensure consistency with original IDL benchmarks and reported critical 5+ bugs.

Grady Healthcare Hospital & Emory Midtown Hospital. Atlanta, GA Intern Aug 2024 – Dec 2024

Conducted contextual inquiry and workflow gap analysis in trauma care; gathered user needs by shadowing clinicians, identifying pain points that informed conceptual device innovations as part of a user-centered design course. NeoDocs Healthcare (YC ’21) Mumbai, Maharashtra

Research Intern Aug 2023 – May 2024

Contributed to product development of microfluidic Vitamin D detection devices through hands-on prototype testing, validation planning, and iterative reliability improvements using customer feedback. Contributed to UV exposure tracking research by assisting in biomarker identification and analyzing correlations between exposure levels and Vitamin D synthesis potential.

Jaslok Hospital and Research Centre Mumbai, Maharashtra Biomedical Engineering Intern Jun 2022 - Jul 2022

Conducted daily inspections and maintenance of medical equipment across departments like ICU, OR’s, Radiology Cardiology, etc. to ensure consistent optimal performance.

Assisted in repair and calibration of equipment including anesthesia machines, ventilators, imaging equipment (MRI, CT, Ultrasound), etc. to reduce downtime and improve operational efficiency. Provided technical support during surgical procedures and achieved zero equipment malfunctions. Contributed to the successful completion of NABH 2022 Audit by supporting departmental compliance efforts. RESEARCH PROJECTS

Analysis of Perianal Fistulize using histopathology data (Jan 2025 – present) Atlanta, GA Applied Hover-Net to segment and classify nuclei in histopathology images, achieving >95% accuracy (Dice score). Extracted 500+ features, reduced dimensions using PCA, UMAP, t-SNE, and selected top 10 features via correlation methods, improving efficiency by 30%.

Achieved AUCs of 0.993 (RF), 0.965 (LR), and 0.993 (NB) using LOSO, 3-fold, and 5-fold cross-validation. Correlated nuclear features with Perianal Crohn’s Disease (PCD), enabling 92% accurate early diagnosis. 3D-Printed Pediatric Heart Valve (Academic Project) Atlanta, GA Designed and fabricated a 3D-printed, patient-specific pediatric pulmonary valve using DLP bioprinting and shape-memory polymers to accommodate somatic growth.

Performed mechanical testing, CAD modeling, and AI-driven anatomical optimization using CT/MRI imaging data. Applied design control principles, conducted risk analysis (FMEA), and adhered to ISO 13485 and FDA 21 CFR 820 standards for Class III device development.

Gene Regulatory Network Reconstruction and Visualization using Machine Learning Atlanta, GA Developed a hybrid GRN model (GENIE3 + Mutual Information) that improved global AUROC by 18.6% (0.89 vs. 0.75) and AUPRC by 23% (0.67 vs. 0.55) over baseline Pearson correlation on 100MR datasets. Optimized hyperparameters (α, bin size) via grid search, achieving 5% higher sensitivity (0.64 vs. 0.49) and 22.3% higher specificity (0.96 vs. 0.79) in TF-target predictions. Validated robustness with 5-fold cross-validation (mean AUROC: 0.86 0.04) and noise tests, retaining >95% accuracy under Gaussian noise (σ=0.2) and ChIP-seq priors.

Automating Skin Allergy Detection Test

M.I.C.E Lab, Sir J.J. Hospital Mumbai, Maharashtra Enhanced diagnostic precision by 40% through advanced image processing techniques using Python and imutils. Increased data processing efficiency by 40% through streamlined data management and real-time analysis with Streamlit. Conducted detailed chart reviews and data extraction for healthcare research projects, improving project outcomes under Dr. Arif Ahmed’s guidance, reducing diagnosis time by 33%.

Developed a machine learning model for medical image classification, achieving 85% accuracy and reducing manual review time by 25%. PUBLICATIONS

Shah, K. T., Nagare, G., Barage, A., Maluskar, A. (2024). "Utilizing Diverse Machine Learning Models for Liver Disease Patient Prediction." 8th International Conference on Computing, Communication, Control, and Automation (ICCUBEA), IEEE. Barage, A, Shah, K., Nagare, G.D., Gidaye, G., 2023. A systematic approach to optimize feature extraction technique for COVID19. Presented at the 7th International Conference on Computing, Communication, Control and Automation.

Contact this candidate