使用從幅度圖像中訓練的生成式先驗模型,並結合相位增強技術,可以顯著提高 MRI 圖像的重建質量,尤其是在高度欠採樣的情況下。
本稿では、位相情報を含まないマグニチュード画像のみを用いて、高精度かつロバストなMRI画像再構成を実現する生成的な事前分布の構築手法を提案する。
本研究提出了一種名為「鄰近切片 Noise2Noise」(NS-N2N)的自我監督式醫學影像去噪方法,僅需使用單一含噪影像體積即可進行訓練,並有效去除醫學影像中的噪聲。
NS-N2N은 단일 노이즈 의료 영상 볼륨에서 인접 슬라이스 간의 유사성을 활용하여 자기 지도 학습을 통해 고품질의 노이즈 제거를 달성하는 새로운 방법이다.
本稿では、単一のノイズを含む医療画像ボリュームのみを用いて、高品質なノイズ除去を実現する新しい自己教師あり学習手法「Neighboring Slice Noise2Noise (NS-N2N)」を提案する。
NS-N2N is a novel self-supervised method that effectively denoises medical images using only a single noisy image volume by leveraging the inherent spatial continuity between neighboring slices to overcome limitations of previous methods reliant on pixel-wise noise independence.
알츠하이머병 환자의 백질 변화 연구에서 2단계 등록 방법을 사용하면 종단 Fixel 기반 분석의 변동성을 줄여 통계적 검정력을 높일 수 있다.
2段階レジストレーション法を用いることで、経時的フィクセルベース解析における測定のばらつきを低減し、統計的検出力を向上させることができる。
A two-step registration method in longitudinal fixel-based analysis (FBA) of diffusion MRI data reduces variability and enhances statistical power for detecting white matter changes in Alzheimer's disease, outperforming the conventional direct registration method.
MRSegmentator is a novel deep learning model that accurately and robustly segments 40 anatomical structures in both MRI and CT images, addressing the limitations of existing organ-specific approaches and offering a valuable tool for automated multi-organ segmentation in medical imaging research.