Decoupled Gradient Harmonized Detector for Partial Annotation: Application to Signet Ring Cell Detection
This work addresses the challenge of automated detection for signet ring cell carcinoma, which is crucial for early diagnosis and improving patient survival rates, though it is incremental in handling label noise in a specific medical domain.
The paper tackled the problem of detecting signet ring cells in medical images under partial annotation and class imbalance, proposing a Decoupled Gradient Harmonizing Mechanism (DGHM-C loss) that achieved second place in the MICCAI DigestPath2019 challenge and showed substantial improvements in controlled experiments.
Early diagnosis of signet ring cell carcinoma dramatically improves the survival rate of patients. Due to lack of public dataset and expert-level annotations, automatic detection on signet ring cell (SRC) has not been thoroughly investigated. In MICCAI DigestPath2019 challenge, apart from foreground (SRC region)-background (normal tissue area) class imbalance, SRCs are partially annotated due to costly medical image annotation, which introduces extra label noise. To address the issues simultaneously, we propose Decoupled Gradient Harmonizing Mechanism (DGHM) and embed it into classification loss, denoted as DGHM-C loss. Specifically, besides positive (SRCs) and negative (normal tissues) examples, we further decouple noisy examples from clean examples and harmonize the corresponding gradient distributions in classification respectively. Without whistles and bells, we achieved the 2nd place in the challenge. Ablation studies and controlled label missing rate experiments demonstrate that DGHM-C loss can bring substantial improvement in partially annotated object detection.