# PA-Seg: Learning from Point Annotations for 3D Medical Image Segmentation using Contextual Regularization and Cross Knowledge Distillation

arXiv: https://arxiv.org/abs/2208.05669 · HTML: https://mlsanity.com/paper/2208.05669/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 46 of its 64 references that are in the map.
- First posted on arXiv 11 Aug 2022.

## Builds on (most cited first, 30 of 46)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels](https://mlsanity.com/paper/1804.06872/) (2018)
- [Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels](https://mlsanity.com/paper/1805.07836/) (2018)
- [When Does Label Smoothing Help?](https://mlsanity.com/paper/1906.02629/) (2019)
- [Learning to Reweight Examples for Robust Deep Learning](https://mlsanity.com/paper/1803.09050/) (2018)
- [Deep Mutual Learning](https://mlsanity.com/paper/1706.00384/) (2017)
- [Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision](https://mlsanity.com/paper/2106.01226/) (2021)
- [Robust Loss Functions under Label Noise for Deep Neural Networks](https://mlsanity.com/paper/1712.09482/) (2017)
- [Uncertainty-aware Self-ensembling Model for Semi-supervised 3D Left Atrium Segmentation](https://mlsanity.com/paper/1907.07034/) (2019)
- [ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1604.05144/) (2016)
- [Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2004.04581/) (2020)
- [Combating noisy labels by agreement: A joint training method with co-regularization](https://mlsanity.com/paper/2003.02752/) (2020)
- [Attention Gated Networks: Learning to Leverage Salient Regions in Medical Images](https://mlsanity.com/paper/1808.08114/) (2018)
- [Semi-supervised Medical Image Segmentation through Dual-task Consistency](https://mlsanity.com/paper/2009.04448/) (2020)
- [Does label smoothing mitigate label noise?](https://mlsanity.com/paper/2003.02819/) (2020)
- [Deep learning with noisy labels: exploring techniques and remedies in medical image analysis](https://mlsanity.com/paper/1912.02911/) (2019)
- [Learning from Noisy Labels with Distillation](https://mlsanity.com/paper/1703.02391/) (2017)
- [On Regularized Losses for Weakly-supervised CNN Segmentation](https://mlsanity.com/paper/1803.09569/) (2018)
- [Normalized Cut Loss for Weakly-supervised CNN Segmentation](https://mlsanity.com/paper/1804.01346/) (2018)
- [DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentation](https://mlsanity.com/paper/1707.00652/) (2017)
- [Unpaired Multi-modal Segmentation via Knowledge Distillation](https://mlsanity.com/paper/2001.03111/) (2020)
- [DeepCut: Object Segmentation from Bounding Box Annotations using Convolutional Neural Networks](https://mlsanity.com/paper/1605.07866/) (2016)
- [Scribble-Supervised Medical Image Segmentation via Dual-Branch Network and Dynamically Mixed Pseudo Labels Supervision](https://mlsanity.com/paper/2203.02106/) (2022)
- [Constrained-CNN losses for weakly supervised segmentation](https://mlsanity.com/paper/1805.04628/) (2018)
- [Inconsistency-aware Uncertainty Estimation for Semi-supervised Medical Image Segmentation](https://mlsanity.com/paper/2110.08762/) (2021)
- [Weakly Supervised Deep Nuclei Segmentation Using Partial Points Annotation in Histopathology Images](https://mlsanity.com/paper/2007.05448/) (2020)

## Built on by (most cited first, 17 of 17)

- [Label-Efficient Deep Learning in Medical Image Analysis: Challenges and Future Directions](https://mlsanity.com/paper/2303.12484/) (2023)
- [Ultrasound Nodule Segmentation Using Asymmetric Learning with Simple Clinical Annotation](https://mlsanity.com/paper/2404.14852/) (2024)
- [Beyond Full Labels: Energy-Double-Guided Single-Point Prompt for Infrared Small Target Label Generation](https://mlsanity.com/paper/2408.08191/) (2024)
- [SAM-Driven Weakly Supervised Nodule Segmentation with Uncertainty-Aware Cross Teaching](https://mlsanity.com/paper/2407.13553/) (2024)
- [Beyond Pixel-Wise Supervision for Medical Image Segmentation: From Traditional Models to Foundation Models](https://mlsanity.com/paper/2404.13239/) (2024)
- [Modality-Agnostic Learning for Medical Image Segmentation Using Multi-modality Self-distillation](https://mlsanity.com/paper/2306.03730/) (2023)
- [Partial Vessels Annotation-based Coronary Artery Segmentation with Self-training and Prototype Learning](https://mlsanity.com/paper/2307.04472/) (2023)
- [Multiple Teachers-Meticulous Student: A Domain Adaptive Meta-Knowledge Distillation Model for Medical Image Classification](https://mlsanity.com/paper/2403.11226/) (2024)
- [Weakly-Supervised Learning via Multi-Lateral Decoder Branching for Tool Segmentation in Robot-Assisted Cardiovascular Catheterization](https://mlsanity.com/paper/2404.07594/) (2024)
- [Weakly Supervised Lymph Nodes Segmentation Based on Partial Instance Annotations with Pre-trained Dual-branch Network and Pseudo Label Learning](https://mlsanity.com/paper/2408.09411/) (2024)
- [Leveraging Different Learning Styles for Improved Knowledge Distillation in Biomedical Imaging](https://mlsanity.com/paper/2212.02931/) (2022)
- [AttenScribble: Attentive Similarity Learning for Scribble-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2312.06614/) (2023)
- [Skeleton Supervised Airway Segmentation](https://mlsanity.com/paper/2403.06510/) (2024)
- [Weakly Supervised Segmentation Framework for Thyroid Nodule Based on High-confidence Labels and High-rationality Losses](https://mlsanity.com/paper/2502.19707/) (2025)
- [CmFNet: Cross-modal Fusion Network for Weakly-supervised Segmentation of Medical Images](https://mlsanity.com/paper/2506.18042/) (2025)
- [RefineSeg: Dual Coarse-to-Fine Learning for Medical Image Segmentation](https://mlsanity.com/paper/2508.02844/) (2025)
- [Prompt-Conditioned Channel Attention for Hierarchical Feature Modulation toward Anatomy-Agnostic Segmentation](https://mlsanity.com/paper/2608.20229/) (2026)

## Abstract (arXiv)

The success of Convolutional Neural Networks (CNNs) in 3D medical image segmentation relies on massive fully annotated 3D volumes for training that are time-consuming and labor-intensive to acquire. In this paper, we propose to annotate a segmentation target with only seven points in 3D medical images, and design a two-stage weakly supervised learning framework PA-Seg. In the first stage, we employ geodesic distance transform to expand the seed points to provide more supervision signal. To further deal with unannotated image regions during training, we propose two contextual regularization strategies, i.e., multi-view Conditional Random Field (mCRF) loss and Variance Minimization (VM) loss, where the first one encourages pixels with similar features to have consistent labels, and the second one minimizes the intensity variance for the segmented foreground and background, respectively. In the second stage, we use predictions obtained by the model pre-trained in the first stage as pseudo labels. To overcome noises in the pseudo labels, we introduce a Self and Cross Monitoring (SCM) strategy, which combines self-training with Cross Knowledge Distillation (CKD) between a primary model and an auxiliary model that learn from soft labels generated by each other. Experiments on public datasets for Vestibular Schwannoma (VS) segmentation and Brain Tumor Segmentation (BraTS) demonstrated that our model trained in the first stage outperformed existing state-of-the-art weakly supervised approaches by a large margin, and after using SCM for additional training, the model's performance was close to its fully supervised counterpart on the BraTS dataset.