Dashuai M.

Key Skills

Software

Other

LabelImg

Label Studio

Labelbox

Top Subject Matter

AI Medical Imaging

Tracheal Cartilage Ultrasound

LLM Medical Hallucination Validation

Top Data Types

Text

Image

Video

Top Task Types

Segmentation

Red Teaming

Freelancer Overview

Project Lead, Medical Image Annotation for CSDA-UNet. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Master of Science, University of Shanghai for Science and Technology (2023). AI-training focus includes data types such as Medical, DICOM, and Text and labeling workflows including Segmentation and Red Teaming.

Labeling Experience

LLM Red Team Medical Hallucination Validator

OtherTextRed Teaming

I designed and executed LLM red team test cases to validate model reliability in interdisciplinary medical, biological, and engineering contexts. The tasks included identifying and labeling hallucinations, logic errors, and domain-specific inconsistencies in LLM outputs. Comprehensive prompt refinement and output evaluation ensured robust assessment of LLM responses. • Tested and labeled LLM medical outputs for hallucinations. • Developed and applied targeted prompts and constraints. • Assessed reasoning, terminology, and causal interpretations. • Focused on specialized medical scenarios in LLM response analysis.

2024 - Present

Project Lead, Medical Image Annotation for CSDA-UNet

OtherSegmentation

I performed detailed pixel-level annotation on emergency airway ultrasound images, using LabelMe to generate segmentation masks for tracheal cartilage. The workflow included data curation, quality control, ROI cropping, edge identification, denoising, and normalization prior to model training. Three-stage quality control ensured high annotation reliability for deep learning use. • Led segmentation labeling for medical imaging AI. • Performed mask generation and advanced preprocessing. • Utilized a rigorous quality control pipeline. • Focused on medical ultrasound for airway emergencies.

2023 - 2023

Education

Work History