Your mugshot

Junya Sato

taich.brave

  •  Japan
  •  Osaka University Graduate School of Medicine
  •  Radiology, Department of Artificial Intelligence Diagnostic Radiology
  •  Website
Statistics
  • Member for 6 years, 10 months
  • 129 challenge submissions
  • 55 algorithms run

Activity Overview

LUNA16 Logo
LUNA16
Challenge User

The LUNA16 challenge: automatic nodule detection on chest CT

CAMELYON17 Logo
CAMELYON17
Challenge User

Automated detection and classification of breast cancer metastases in whole-slide images of histological lymph node sections. This task has high clinical relevance and would normally require extensive microscopic assessment by pathologists.

MELA Logo
MELA2022
Challenge User

MICCAI 2022 MELA Challenge: A Large-Scale Detection Benchmark of 1,100 CT Scans for Mediastinal Lesion Analysis

Parse2022 Logo
Parse2022
Challenge User

It is of significant clinical interest to study pulmonary artery structures in the field of medical image analysis. One prerequisite step is to segment pulmonary artery structures from CT with high accuracy and low time-consuming. The segmentation of pulmonary artery structures benefits the quantification of its morphological changes for diagnosis of pulmonary hypertension and thoracic surgery. However, due to the complexity of pulmonary artery topology, automated segmentation of pulmonary artery topology is a challenging task. Besides, the open accessible large-scale CT data with well labeled pulmonary artery are scarce (The large variations of the topological structures from different patients make the annotation an extremely challenging process). The lack of well labeled pulmonary artery hinders the development of automatic pulmonary artery segmentation algorithm. Hence, we try to host the first Pulmonary ARtery SEgmentation challenge in MICCAI 2022 (Named Parse2022) to start a new research topic.

autoPET Logo
autoPET
Challenge User

Automatic lesion segmentation in whole-body FDG-PET/CT

AMOS22 Logo
Multi-Modality Abdominal Multi-Organ Segmentation Challenge 2022
Challenge User

ATM22 Logo
Multi-site, Multi-Domain Airway Tree Modeling (ATM’22)
Challenge User

Airway segmentation is a crucial step for the analysis of pulmonary diseases including asthma, bronchiectasis, and emphysema. The accurate segmentation based on X-Ray computed tomography (CT) enables the quantitative measurements of airway dimensions and wall thickness, which can reveal the abnormality of patients with chronic obstructive pulmonary disease (COPD). Besides, the extraction of patient-specific airway models from CT images is required for navigatiisted surgery.

HaN-Seg2023 Logo
The Head and Neck Organ-at-Risk CT & MR Segmentation Challenge
Challenge User

A semantic multimodal segmentation challenge comprising 30 organs at risk

autoPET-II Logo
autoPET-II
Challenge User

Automated Lesion Segmentation in PET/CT - Domain Generalization

toothfairy Logo
ToothFairy: Cone-Beam Computed Tomography Segmentation Challenge
Challenge User

This is the first edition of the ToothFairy challenge organized by the University of Modena and Reggio Emilia with the collaboration of Raudboud University. This challenge aims at pushing the development of deep learning frameworks to segment the Inferior Alveolar Canal (IAC) by incrementally extending the amount of publicly available 3D-annotated Cone Beam Computed Tomography (CBCT) scans. CBCT modality is becoming increasingly important for treatment planning and diagnosis in implant dentistry and maxillofacial surgery. The three-dimensional information acquired with CBCT can be crucial to plan a vast number of surgical interventions with the aim of preserving noble anatomical structures such as the Inferior Alveolar Canal (IAC), which contains the homonymous nerve (Inferior Alveolar Nerve, IAN). Deep learning models can support medical personnel in surgical planning procedures by providing a voxel-level segmentation of the IAN automatically extracted from CBCT scans.

LNQ2023 Logo
LNQ2023
Challenge User

Accurate lymph node size estimation is critical for staging cancer patients, initial therapeutic management, and in longitudinal scans, assessing response to therapy. Current standard practice for quantifying lymph node size is based on a variety of criteria that use unidirectional or bidirectional measurements on just one or a few nodes, typically on just one axial slice. But humans have hundreds of lymph nodes, any number of which may be enlarged to various degrees due to disease or immune response. While a normal lymph node may be approximately 5mm in diameter, a diseased lymph node may be several cm in diameter. The mediastinum, the anatomical area between the lungs and around the heart, may contain ten or more lymph nodes, often with three or more enlarged greater than 1cm. Accurate segmentation in 3D would provide more information to evaluate lymph node disease.

MultiCenterAorta Logo
SEG.A. - Segmentation of the Aorta
Challenge User

Segmentation, modeling and visualization of the arterial tree are still a challenge in medical image analysis. The main track of this challenge deals with the fully automatic segmentation of the aortic vessel tree in computed tomography images. Optionally, teams can submit tailored solutions for meshing and visualization of the vessel tree.

SPPIN Logo
Surgical Planning in Pediatric Neuroblastoma
Challenge User

SegRap2023 Logo
SegRap 2023
Challenge User

A segmentation challenge with 200 patients (two modalities of CT images, 45 OARs and 2 GTVs).

MONKEY Logo
MONKEY challenge: Detection of inflammation in kidney biopsies
Challenge User

MONKEY (Machine-learning for Optimal detection of iNflammatory cells in KidnEY)

PUMA Logo
PUMA: Panoptic segmentation of nUclei and tissue in MelanomA
Challenge User

The PUMA Challenge aims to enhance nuclei and tissue segmentation in melanoma histopathology, addressing the need for better prognostic biomarkers to predict treatment responses. Melanoma, a highly aggressive skin cancer, often requires immune checkpoint inhibition therapy, but only half of patients respond. Prognostic biomarkers like tumor infiltrating lymphocytes (TILs) correlate with better therapy responses and lower recurrence rate, but manual TIL scoring is subjective and inconsistent. Current deep learning methods underperform. The PUMA dataset includes annotated primary and metastatic melanoma regions to improve segmentation techniques. The challenge includes two tracks with tasks focused on tissue and nuclei segmentation, encouraging advanced methods to improve predictive accuracy.