Amalie Shi

Amalie Shi

Amalie Shi is a biomedical and software engineer specializing in healthcare automation and medical imaging verification. With degrees from Johns Hopkins University and Georgia Tech, she integrates Python-based frameworks to improve clinical software reliability. A certified PMP, EIT, ISTQB CTAL-TAE, and ASQ CSSYB, she advances innovation in digital healthcare systems.

Posts by Collection

portfolio

UI Software Testing Framework

Published:

An automated UI software testing framework for medical imaging applications, presented at RSLondon Southeast and awarded Best Poster.

For more information, click: here

publications

Positive Force Feedback May Ameliorate Muscle Weakness

Published in Society for Neuroscience Annual Meeting 2017, 2017

Abstract presented at the Society for Neuroscience (SfN) Annual Meeting 2017. Investigates whether positive force feedback from Golgi tendon organs can enhance load-bearing tasks and ameliorate muscle weakness via feedback-controlled electrical stimulation.

For more information, click: here

Performance Assessment of Texture Reproduction in High-Resolution CT

Published in SPIE, 2020

Assessment of computed tomography (CT) images can be complex due to a number of dependencies that affect system performance. In particular, it is well-known that noise in CT is object-dependent. Such object-dependence can be more pronounced and extend to resolution and image textures with the increasing adoption of model-based reconstruction and processing with machine learning methods. Moreover, such processing is often inherently nonlinear complicating assessments with simple measures of spatial resolution, etc. Similarly, recent advances in CT system design have attempted to improve fine resolution details - e.g., with newer detectors, smaller focal spots, etc. Recognizing these trends, there is a greater need for imaging assessment that are considering specific features of interest that can be placed within an anthropomorphic phantom for realistic emulation and evaluation. In this work, we devise a methodology for 3D-printing phantom inserts using procedural texture generation for evaluation of performance of high-resolution CT systems. Accurate representations of texture have previously been a hindrance to adoption of processing methods like model-based reconstruction, and texture serves as an important diagnostic feature (e.g. heterogeneity of lesions is a marker for malignancy). We consider the ability of different systems to reproduce various textures (as a function of the intrinsic feature sizes of the texture), comparing microCT, cone-beam CT, and diagnostic CT using normal- and high-resolution modes. We expect that this general methodology will provide a pathway for repeatable and robust assessments of different imaging systems and processing methods.

For more information, click: here

Building Scalable and Automated Verification Tools for Healthcare Imaging Systems

Published in MIE 2026 Conference, 2026

Traditional verification of healthcare imaging systems remains burdened by manual effort, fragmented vendor implementations, and limited traceability. Ensuring compliance with DICOM conformance statements across interacting components often requires testers to repeat labour-intensive checks across multiple systems, configurations, and protocol variants. As standards evolve and legacy infrastructures are retired, these activities struggle to keep pace with modern release cycles and become increasingly unsustainable, error-prone, and difficult to scale. This paper presents a human-centered automation framework that modernizes healthcare imaging verification through a unified pipeline of automated tools spanning key software subsystems. The framework abstracts underlying tool complexity and enables engineers and QA professionals to collaboratively execute, monitor, and interpret verification tasks. Guided by human-factor design principles, the approach lowers cognitive overhead, streamlines reporting, and improves transparency through repeatable workflows and traceable verification artifacts. The result is a scalable, adaptive, and user-friendly verification process that bridges standards-driven compliance with modern software engineering practices.

For more information, click: here

talks

Positive Feedback Control in Neural Systems

Published:

Functional electrical stimulation can activate weakened or paralyzed muscles during locomotion by using kinematic signals to control stimulation. Building on evidence that positive force feedback from Golgi tendon organs enhances load-bearing tasks, we engineered a feedback-controlled stimulation system to increase muscle force in cases of muscle weakness. In decerebrate cats, muscle force was recorded during stretch and used to trigger intramuscular electrical stimulation, amplifying force output beyond the stretch reflex. Adjustable thresholds and gains allowed us to explore system stability, revealing that stimulation remained stable across a wide parameter range, but excessive gain or low thresholds led to prolonged contractions. Clinically, electromyographic signals could provide feedback, with stimulation delivered via intramuscular or surface electrodes.

UI Software Testing Framework

Published:

An automated UI software testing framework for medical imaging applications, presented at RSLondon

teaching

Calculus 1 & 2 Teaching Assistant

Undergraduate course, Georgia Institute of Technology, Department of Mathematics, 2013

Served as a Teaching Assistant for Calculus 1 and 2 from 2013 to 2016, leading recitation sessions to help students master derivatives, integration, and advanced calculus concepts. Provided guidance, answered questions, and facilitated problem-solving to enhance student understanding and performance.