# Densely Decoded Networks with Adaptive Deep Supervision for Medical Image Segmentation

arXiv: https://arxiv.org/abs/2402.02649 · HTML: https://mlsanity.com/paper/2402.02649/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 31 references that are in the map.
- First posted on arXiv 5 Feb 2024.

## Builds on (most cited first, 23 of 23)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/1606.04797/) (2016)
- [Understanding the Effective Receptive Field in Deep Convolutional Neural Networks](https://mlsanity.com/paper/1701.04128/) (2017)
- [Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)](https://mlsanity.com/paper/1710.05006/) (2017)
- [Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge](https://mlsanity.com/paper/1811.02629/) (2018)
- [Deeply-Supervised Nets](https://mlsanity.com/paper/1409.5185/) (2014)
- [Brain Tumor Segmentation and Radiomics Survival Prediction: Contribution to the BRATS 2017 Challenge](https://mlsanity.com/paper/1802.10508/) (2018)
- [Automatic Brain Tumor Detection and Segmentation Using U-Net Based Fully Convolutional Networks](https://mlsanity.com/paper/1705.03820/) (2017)
- [Automatic 3D Cardiovascular MR Segmentation with Densely-Connected Volumetric ConvNets](https://mlsanity.com/paper/1708.00573/) (2017)
- [Skin Lesion Analysis Towards Melanoma Detection Using Deep Learning Network](https://mlsanity.com/paper/1703.00577/) (2017)
- [Improving Dermoscopic Image Segmentation with Enhanced Convolutional-Deconvolutional Networks](https://mlsanity.com/paper/1709.09780/) (2017)
- [CC-Net: Image Complexity Guided Network Compression for Biomedical Image Segmentation](https://mlsanity.com/paper/1901.01578/) (2019)
- [Solo or Ensemble? Choosing a CNN Architecture for Melanoma Classification](https://mlsanity.com/paper/1904.12724/) (2019)
- [Data-Driven Deep Supervision for Skin Lesion Classification](https://mlsanity.com/paper/2209.01527/) (2022)
- [Decompose-and-Integrate Learning for Multi-class Segmentation in Medical Images](https://mlsanity.com/paper/1906.02901/) (2019)
- [Objective-Dependent Uncertainty Driven Retinal Vessel Segmentation](https://mlsanity.com/paper/2104.08554/) (2021)
- [A New Registration Approach for Dynamic Analysis of Calcium Signals in Organs](https://mlsanity.com/paper/1802.00491/) (2018)
- [Towards Automated Melanoma Detection with Deep Learning: Data Purification and Augmentation](https://mlsanity.com/paper/1902.06061/) (2019)

## Abstract (arXiv)

Medical image segmentation using deep neural networks has been highly successful. However, the effectiveness of these networks is often limited by inadequate dense prediction and inability to extract robust features. To achieve refined dense prediction, we propose densely decoded networks (ddn), by selectively introducing 'crutch' network connections. Such 'crutch' connections in each upsampling stage of the network decoder (1) enhance target localization by incorporating high resolution features from the encoder, and (2) improve segmentation by facilitating multi-stage contextual information flow. Further, we present a training strategy based on adaptive deep supervision (ads), which exploits and adapts specific attributes of input dataset, for robust feature extraction. In particular, ads strategically locates and deploys auxiliary supervision, by matching the average input object size with the layer-wise effective receptive fields (lerf) of a network, resulting in a class of ddns. Such inclusion of 'companion objective' from a specific hidden layer, helps the model pay close attention to some distinct input-dependent features, which the network might otherwise 'ignore' during training. Our new networks and training strategy are validated on 4 diverse datasets of different modalities, demonstrating their effectiveness.