For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
Hirozumi Mori

Hirozumi Mori

Primary Data Annotator and AI Model Contributor

JAPAN flag
Tokyo, Japan
$150.00/hrExpertOther

Key Skills

Software

Other

Top Subject Matter

Medical Imaging
Deep Learning
Adrenal Vein Sampling

Top Data Types

ImageImage
Medical DicomMedical Dicom

Top Task Types

Segmentation
Classification

Freelancer Overview

Primary Data Annotator and AI Model Contributor. Brings 16+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Doctor of Medicine, The University of Tokyo (2011) and Doctor of Philosophy, Keio University (2023). AI-training focus includes data types such as Medical and DICOM and labeling workflows including Segmentation.

ExpertEnglish

Labeling Experience

Medical Imaging AI Annotation & Quality Assurance Specialist (Radiologist)

Medical DicomSegmentation
Performed high-level medical image annotation and quality assessment for AI model development at LPIXEL Inc. over a period exceeding two years. Contributed to multiple clinical domains, including: Cerebral infarction (MRI-based lesion detection and segmentation) Lung nodule detection (CT-based identification and characterization) Ischemia evaluation (multi-modality imaging interpretation) Key responsibilities included: Precise lesion annotation and segmentation on CT and MRI datasets Image-level and pixel-level labeling for supervised learning models Comprehensive quality assurance (QA) and validation of annotated datasets Development and refinement of gold-standard annotation guidelines to ensure consistency across annotators Collaboration with AI engineers and data scientists to improve model performance and clinical relevance Project scale involved large, multi-case imaging datasets, requiring strict adherence to clinical accuracy and reproducibility standards. All annotations were performed in accordance with radiological best practices, ensuring high-quality training data suitable for advanced AI model development.

Performed high-level medical image annotation and quality assessment for AI model development at LPIXEL Inc. over a period exceeding two years. Contributed to multiple clinical domains, including: Cerebral infarction (MRI-based lesion detection and segmentation) Lung nodule detection (CT-based identification and characterization) Ischemia evaluation (multi-modality imaging interpretation) Key responsibilities included: Precise lesion annotation and segmentation on CT and MRI datasets Image-level and pixel-level labeling for supervised learning models Comprehensive quality assurance (QA) and validation of annotated datasets Development and refinement of gold-standard annotation guidelines to ensure consistency across annotators Collaboration with AI engineers and data scientists to improve model performance and clinical relevance Project scale involved large, multi-case imaging datasets, requiring strict adherence to clinical accuracy and reproducibility standards. All annotations were performed in accordance with radiological best practices, ensuring high-quality training data suitable for advanced AI model development.

2024 - Present

Primary Data Annotator and AI Model Contributor

OtherMedical DicomSegmentation
Participated as the primary contributor in a JSPS-funded AI research project focused on developing a 3D U-Net segmentation model for automatic detection of the right adrenal vein in medical imaging. Main responsibilities included data annotation, preprocessing, and implementation of the AI model using Python. Ensured high-quality segmentation labels and collaborated with a multidisciplinary team to optimize model performance. • Annotated DICOM medical images for accurate right adrenal vein identification • Developed and implemented preprocessing pipelines for model training • Validated label accuracy to support robust AI development • Collaborated on improving model generalization across imaging cases

Participated as the primary contributor in a JSPS-funded AI research project focused on developing a 3D U-Net segmentation model for automatic detection of the right adrenal vein in medical imaging. Main responsibilities included data annotation, preprocessing, and implementation of the AI model using Python. Ensured high-quality segmentation labels and collaborated with a multidisciplinary team to optimize model performance. • Annotated DICOM medical images for accurate right adrenal vein identification • Developed and implemented preprocessing pipelines for model training • Validated label accuracy to support robust AI development • Collaborated on improving model generalization across imaging cases

2024 - 2024

Education

T

The University of Tokyo

Doctor of Medicine, Medicine

Doctor of Medicine
2004 - 2011
K

Keio University

Doctor of Philosophy, Diagnostic Radiology

Doctor of Philosophy
2023

Work History

K

Keio University

Radiology Fellow

Tokyo
2023 - Present
S

Saitama City Hospital

Radiology Resident

Saitama
2021 - 2023