10 papers
XTinyU-Net: Training-Free U-Net Scaling via Initialization-Time Sensitivity
Alvin Kimbowa, Moein Heidari, David Liu +1
While U-Net architectures remain the gold standard for medical image segmentation, their deployment in resource-constrained environments demands aggressive model compression. Howev…
When Minor Edits Matter: LLM-Driven Prompt Attack for Medical VLM Robustness in Ultrasound
Yasamin Medghalchi, Milad Yazdani, Amirhossein Dabiriaghdam +7
Ultrasound is widely used in clinical practice due to its portability, cost-effectiveness, safety, and real-time imaging capabilities. However, image acquisition and interpretation…
Echo-CoPilot: A Multiple-Perspective Agentic Framework for Reliable Echocardiography Interpretation
Moein Heidari, Ali Mehrabian, Mohammad Amin Roohi +4
Echocardiography interpretation requires integrating multi-view temporal evidence with quantitative measurements and guideline-grounded reasoning, yet existing foundation-model pip…
WaRA: Wavelet Low Rank Adaptation
Moein Heidari, Yijin Huang, Yasamin Medghalchi +3
Adapting large pretrained vision models to medical image classification is often limited by memory, computation, and task-specific specializations. Parameter-efficient fine-tuning…
Echo-ENet: Efficient Endocardial Spatio-Temporal Network for Ejection Fraction Estimation
Moein Heidari, Afshin Bozorgpour, AmirHossein Zarif-Fakharnia +5
Objective To develop a robust and computationally efficient deep learning model for automated left ventricular ejection fraction (LVEF) estimation from echocardiography videos that…
SLA-INR: Single-Layer Learnable Activation for Implicit Neural Representation
Moein Heidari, Reza Rezaeian, Reza Azad +3
Implicit Neural Representation (INR), leveraging a neural network to transform coordinate input into corresponding attributes, has recently driven significant advances in several v…