Dice Similarity Coefficient

The Dice Similarity Coefficient (DSC) is a widely used metric for evaluating the accuracy of image segmentations, particularly in medical imaging, aiming to quantify the overlap between a predicted segmentation and a ground truth. Recent research highlights limitations of DSC, including bias towards class prevalence and insensitivity to subtle segmentation variations, leading to investigations into alternative or supplementary metrics like radiomics features and normalized DSC (nDSC). These studies often involve deep learning models, such as convolutional neural networks (CNNs) and transformers, and focus on improving segmentation accuracy and developing more robust evaluation methods for tasks like organ-at-risk delineation in radiotherapy planning. The ultimate goal is to enhance the reliability and clinical utility of automated segmentation techniques in various medical applications.

Papers

October 30, 2023

Radiomics as a measure superior to the Dice similarity coefficient for tumor segmentation performance evaluation
Yoichi Watanabe, Rukhsora Akramova
Tumor Segmentation Radiomics Feature Dice Similarity Coefficient

March 20, 2023

Convolutions, Transformers, and their Ensembles for the Segmentation of Organs at Risk in Radiation Treatment of Cervical Cancer
Vangelis Kostoulas, Peter A. N. Bosman, Tanja Alderliesten
Deep Neural Network Transformer Megatron Decepticons Segmentation Based Approach Direct Convolution Radiation Oncology Cervical Cancer Performing Model Dice Similarity Coefficient

February 10, 2023

Tackling Bias in the Dice Similarity Coefficient: Introducing nDSC for White Matter Lesion Segmentation
Vatsal Raina, Nataliia Molchanova, Mara Graziani, Andrey Malinin, Henning Muller, Meritxell Bach Cuadra, Mark Gales
Absolute Stance Bias Magnetic Resonance Image White Matter Lesion Segmentation Dice Similarity Coefficient

November 8, 2022

Theoretical analysis and experimental validation of volume bias of soft Dice optimized segmentation maps in the context of inherent uncertainty
Jeroen Bertels, David Robben, Dirk Vandermeulen, Paul Suetens
Experimental Validation Automatic Volume Segmentation Map Differentiable Surrogate Dice Similarity Coefficient Inherent Uncertainty

September 25, 2022

Contour Dice loss for structures with Fuzzy and Complex Boundaries in Fetal MRI
Bella Specktor Fadida, Bossmat Yehuda, Daphna Link Sourani, Liat Ben Sira, Dafna Ben Bashat, Leo Joskowicz
Fetal Brain Simple Structure Fetal MRI Placenta Segmentation Dice Loss Dice Similarity Coefficient Uncertain Boundary

March 30, 2022

The impact of using voxel-level segmentation metrics on evaluating multifocal prostate cancer localisation
Wen Yan, Qianye Yang, Tom Syer, Zhe Min, Shonit Punwani, Mark Emberton, Dean C. Barratt, Bernard Chiu, Yipeng Hu
Global Impact Prostate Cancer Voxel Based Semantic Segmentation Hausdorff Distance Lesion Wise Metric Dice Similarity Coefficient

Dice Similarity Coefficient

Papers

Radiomics as a measure superior to the Dice similarity coefficient for tumor segmentation performance evaluation

Convolutions, Transformers, and their Ensembles for the Segmentation of Organs at Risk in Radiation Treatment of Cervical Cancer

Tackling Bias in the Dice Similarity Coefficient: Introducing nDSC for White Matter Lesion Segmentation

Theoretical analysis and experimental validation of volume bias of soft Dice optimized segmentation maps in the context of inherent uncertainty

Contour Dice loss for structures with Fuzzy and Complex Boundaries in Fetal MRI

The impact of using voxel-level segmentation metrics on evaluating multifocal prostate cancer localisation