核心概念
개발된 새로운 알고리즘인 APPLE은 의료 분할 작업에서 사전 훈련된 기본 모델의 아키텍처나 매개변수를 수정하지 않고 모델의 공정성을 향상시킵니다.
統計
Ensuring fairness in deep-learning-based segmentors is crucial for health equity.
Experiments on two segmentation datasets and five segmentors illustrate the effectiveness of APPLE.
Extensive experiments on two medical segmentation datasets prove the effectiveness of APPLE.
引用
"Ensuring fairness in deep-learning-based segmentors is crucial for health equity."
"Experiments on two segmentation datasets and five segmentors illustrate the effectiveness of our proposed method compared to several unfairness mitigation methods."
"Extensive experiments on two medical segmentation datasets prove that APPLE can improve fairness in deployed segmentors effectively."