MEDIAR: Harmony of Data-Centric and Model-Centric for Multi-Modality Microscopy (2022-12-07T00:00:00.000000Z)

TL;DR

MEDIAR harmonizes data-centric and model-centric approaches as the learning and inference strategies, achieving a 0.9067 F1-score at the validation phase while satisfying the time budget.

Abstract

Cell segmentation is a fundamental task for computational biology analysis. Identifying the cell instances is often the first step in various downstream biomedical studies. However, many cell segmentation algorithms, including the recently emerging deep learning-based methods, still show limited generality under the multi-modality environment. Weakly Supervised Cell Segmentation in Multi-modality High-Resolution Microscopy Images was hosted at NeurIPS 2022 to tackle this problem. We propose MEDIAR, a holistic pipeline for cell instance segmentation under multi-modality in this challenge. MEDIAR harmonizes data-centric and model-centric approaches as the learning and inference strategies, achieving a 0.9067 F1-score at the validation phase while satisfying the time budget. To facilitate subsequent research, we provide the source code and trained model as open-source: https://github.com/Lee-Gihun/MEDIAR

Authors

Joonkee Kim

2 papers

Gihun Lee

2 papers

Se-Young Yun

4 papers

TL;DR

Abstract

Authors

References68 items

DetecDiv, a generalist deep-learning platform for automated cell division tracking and survival analysis

Semi-Supervised Video Semantic Segmentation with Inter-Frame Feature Reconstruction

Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images

Generative Adversarial Networks

Masked Autoencoders Are Scalable Vision Learners

Omnipose: a high-precision morphology-independent solution for bacterial cell segmentation

Avoiding a replication crisis in deep-learning-based bioimage analysis

LIVECell—A large-scale dataset for label-free live cell segmentation

Re-distributing Biased Pseudo Labels for Semi-supervised Semantic Segmentation: A Baseline Investigation

MST: Masked Self-Supervised Transformer for Visual Representation

SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers

Image-based cell phenotyping with deep learning.

Emerging Properties in Self-Supervised Vision Transformers

Democratising deep learning for microscopy with ZeroCostDL4Mic

UNETR: Transformers for 3D Medical Image Segmentation

Whole-cell segmentation of tissue images with human-level performance using large-scale data annotation and deep learning

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

PseudoSeg: Designing Pseudo Labels for Semantic Segmentation

MixCo: Mix-up Contrastive Learning for Visual Representation

Deep learning in digital pathology image analysis: a survey

Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models

In Search of Lost Domain Generalization

Unsupervised Learning of Visual Features by Contrasting Cluster Assignments

A Noise-Robust Framework for Automatic Segmentation of COVID-19 Pneumonia Lesions From CT Images

Semi-Supervised Semantic Segmentation With Cross-Consistency Training

Deep Learning in Mining Biological Data

A Simple Framework for Contrastive Learning of Visual Representations

Cellpose: a generalist algorithm for cellular segmentation

PyTorch: An Imperative Style, High-Performance Deep Learning Library

Nucleus segmentation across imaging experiments: the 2018 Data Science Bowl

Using Hindsight to Anchor Past Knowledge in Continual Learning

Mish: A Self Regularized Non-Monotonic Neural Activation Function

Multi-Task Attention-Based Semi-Supervised Learning for Medical Image Segmentation

On Tiny Episodic Memories in Continual Learning

Moment Matching for Multi-Source Domain Adaptation

Multi-Task Learning as Multi-Objective Optimization

A Structured Tumor-Immune Microenvironment in Triple Negative Breast Cancer Revealed by Multiplexed Ion Beam Imaging

A Two-Stream Mutual Attention Network for Semi-Supervised Biomedical Segmentation with Noisy Labels

Cell Detection with Star-convex Polygons

Continual Lifelong Learning with Neural Networks: A Review

Decoupled Weight Decay Regularization

Mixed Precision Training

Deeper, Broader and Artier Domain Generalization

Data-analysis strategies for image-based cell profiling

Attention is All you Need

Mask R-CNN

A survey on deep learning in medical image analysis

SGDR: Stochastic Gradient Descent with Warm Restarts

Microscopy-Based High-Content Screening

U-Net: Convolutional Networks for Biomedical Image Segmentation

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks

The stability-plasticity dilemma: investigating the continuum from catastrophic forgetting to age-limited learning effects

ImageNet classification with deep convolutional neural networks

Cell Segmentation: 50 Years Down the Road [Life Sciences]

Fiji: an open-source platform for biological-image analysis

Ilastik: Interactive learning and segmentation toolkit

ImageNet: A large-scale hierarchical image database

Apoptosis in development

Multitask Learning

MA-Net: A Multi-Scale Attention Network for Liver and Tumor Segmentation

Image test time augmentation with pytorch

Consistency-based Semi-supervised Learning for Object detection

Deep Learning

Segmentation Models

Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem

Calculus with Analytic Geometry

Monai: Medical open network for ai

Field of Study

Journal Information

Name

Volume