An AI-driven system, ReXplain, generates comprehensive video reports that translate complex radiology findings into plain language, highlight relevant anatomical regions, and utilize an avatar to simulate one-on-one consultations, aiming to improve patient understanding and engagement.
RadiomicsFill-Mammo, a novel approach, leverages radiomics features to generate realistic synthetic mammogram masses with desired attributes, enabling data augmentation and enhancing downstream medical imaging tasks.
The AI in Medical Imaging (AIMI) project developed state-of-the-art nnU-Net models to generate accurate annotations for cancer radiology images in the National Cancer Institute's Imaging Data Commons, facilitating further research and development in cancer imaging.
복잡하고 구분하기 어려운 복부 해부학적 구조를 정확하게 재현하는 고품질의 합성 복부 CT 볼륨을 생성하여 자기 지도 학습 기반 장기 분할 성능을 향상시킬 수 있다.
Integrating pre-trained models into a Federated Learning framework can improve diagnostic accuracy and robustness against image corruption in kidney stone identification, addressing privacy concerns and enhancing patient care.
Infrared imaging of the meibomian glands in the eyelids can help detect early signs of Sjögren's disease, a chronic autoimmune disorder that often goes undiagnosed for years.
An automated deep learning method that can accurately reconstruct the 3D model of the liver and estimate its volume using just three partial ultrasound scans, without requiring the full view of the organ.
Segment Anything Model 2 (SAM 2) demonstrated promising zero-shot performance in segmenting certain abdominal organs in CT scans, particularly larger organs with clear boundaries, but struggled with smaller and less defined structures.
자유 호흡 상태에서 상체 자기공명 지문 기법을 이용하여 수분 T1 및 지방 분율을 정량화하고, 호흡 운동에 의한 영향을 보정하는 방법을 제안하였다.
An ensemble of deep learning models, including UNet, ResNet, EfficientNet, and VGG, achieves superior performance in segmenting the left and right atria and their walls from late gadolinium-enhanced cardiac MRI data of atrial fibrillation patients.