MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification.

Jiancheng Yang, Rui Shi, Donglai Wei, Zequan Liu, Lin Zhao, Bilian Ke, Hanspeter Pfister, Bingbing Ni

January 2023 Sci Data

Synopsis of Social media discussions

The overall discussions reflect strong interest and appreciation for the dataset’s scale and openness, evidenced by phrases like 'large-scale lightweight benchmark' and mentions of datasets with hundreds of thousands of images. Words like 'support,' 'potential,' and 'recognize' show a positive tone, emphasizing the tool’s utility for advancing biomedical research, while some comments clarify its non-clinical purpose, which tempers the perceived impact.

A
Agreement
 Moderate agreement

Most discussions recognize the publication’s value as a comprehensive resource for biomedical image classification, with some expressing support for its extensive dataset and open-source approach.

I
Interest
 High level of interest

Posts demonstrate high curiosity, highlighting the large scale of the dataset and its potential use in various research fields.

E
Engagement
 Moderate level of engagement

Several discussions delve into methodological aspects and the significance of providing standardized benchmarks, indicating active engagement.

I
Impact
 Moderate level of impact

The emphasis on broad accessibility and the utility for research suggests a perception of moderate impact, though some comments downplay clinical relevance.

Social Mentions

YouTube

1 Videos

Twitter

27 Posts

Metrics

Video Views

119

Total Likes

47

Extended Reach

114,344

Social Features

28

Timeline: Posts about article

Top Social Media Posts

Posts referencing the article

MedMNIST v2: Large-Scale Benchmark for Biomedical Image Classification

MedMNIST v2: Large-Scale Benchmark for Biomedical Image Classification

MedMNIST v2 is a comprehensive collection of standardized biomedical images, including over 700,000 2D and nearly 10,000 3D images, designed for easy use in classification research and education. All data and code are openly available.

STEM RTCL TV

April 27, 2023

119 views

  • Benjamin Bergner
    @bergbenj (Twitter)

    MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification Datasets with MNIST-like properties, but for medical imaging. Provides 708,069 2D and 9,998 3D images across 12 2D and 6 3D datasets. https://t.co/iuHMs3Mqg9 https://t.co/1NNHJx29BF
    view full post

    February 9, 2023

    1

  • Paul Lopez
    @lopezunwired (Twitter)

    MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification https://t.co/lffrZRERPW #MachineLearning #NatureJournal #AI https://t.co/ayqdxmOFYq
    view full post

    January 19, 2023

    1

  • Reluctant Quant
    @DrMattCrowson (Twitter)

    RT MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D biomedical image classification https://t.co/a2ureL4MoR https://t.co/Mmlt3z1Za2
    view full post

    January 19, 2023

  • Egutz
    @Egutz_ (Twitter)

    MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D ... - https://t.co/Ik4B0H1LnA https://t.co/wG7ovEmxLJ https://t.co/FER4AgBlz4
    view full post

    January 19, 2023

  • Philippe JEAN-BAPTISTE
    @PhilippeJB_PJB (Twitter)

    MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D ... - https://t.co/HPXr7mQKM8 https://t.co/YjXAUv271e
    view full post

    January 19, 2023

  • Richard Hill
    @profrhill (Twitter)

    MedMNIST v2 - A large-scale lightweight benchmark for 2D and 3D ... - https://t.co/I9OYexdPXs https://t.co/wZ0wwL60mv
    view full post

    January 19, 2023

  • arXiv reaDer bot (cs-CV)
    @arXiv_reaDer (Twitter)

    MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification MedMNIST v2 -- 2D および 3D 生物医学画像分類のための大規模で軽量なベンチマーク 2022-09-25T06:07:53+00:00 arXiv: https://t.co/SsKllBjxpB 英/日サマリ↓ https://t.co/JHZzRT4KUx
    view full post

    September 26, 2022

    1

  • SkinHelpDesk.com
    @SkinHelpDesk (Twitter)

    #MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical #ImageClassification https://t.co/98uM563fNL via @beapen
    view full post

    December 3, 2021

  • Bell Eapen
    @beapen (Twitter)

    #MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical #ImageClassification https://t.co/t1xKYQcpMA
    view full post

    December 3, 2021

  • HubOfML
    @hubofml (Twitter)

    RT @Deep__AI: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/Iq6XrbDwVq by Jia…
    view full post

    November 14, 2021

    2

  • OpenSource Orgs
    @opensource_orgs (Twitter)

    RT @Deep__AI: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/Iq6XrbDwVq by Jia…
    view full post

    November 14, 2021

    2

  • DeepAI
    @DeepAI (Twitter)

    MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/Iq6XrbDwVq by Jiancheng Yang et al. #ComputerVision #OpenSource
    view full post

    October 30, 2021

    3

    2

  • Konstantinos Kyriakidis
    @kokyriakidis (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 30, 2021

    10

  • GOTRUVAY
    @gotruvay (Twitter)

    MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/ClbFdiITxz Become a better expert via https://t.co/3Lriam4BWS #AI #MachineLearning #ArtificialIntelligence #NeuralNetworks #100DaysOfCode #Robotics
    view full post

    October 30, 2021

  • 蒼月@かずにゃん
    @kazunyan_nyan (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 30, 2021

    10

  • Steven Edwards
    @stephenwithavee (Twitter)

    MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/af6KjhhZzU
    view full post

    October 29, 2021

  • ぎゃばん＠にーくらのハリセン。
    @gavangavan (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • priya joseph
    @ayirpelle (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • tavo
    @TavoGLC (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • levan lev
    @dr_levan (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • Khanh Nguyen
    @The_Anh_Ta (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • Muktabh Mayank
    @muktabh (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • 午後のarXiv
    @arxivml (Twitter)

    "MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification", Jiancheng Yang, R… https://t.co/3uwZ2e34wp
    view full post

    October 29, 2021

  • Siladittya Manna
    @sadimanna (Twitter)

    RT @ducha_aiki: MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;d…
    view full post

    October 29, 2021

    10

  • Dmytro Mishkin
    @ducha_aiki (Twitter)

    MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification https://t.co/7tZODHi3gb tl;dr: Bored of fashionMNIST? P.S. NOT intended for clinical use. Jiancheng Yang,Rui Shi,Donglai Wei,Zequan Liu,Lin Zhao,Bilian Ke,Hanspeter Pfister,Bingbing Ni https://t.co/eR3ugjPTsj
    view full post

    October 29, 2021

    40

    10

  • Tim Leiner
    @MLandDL_papers (Twitter)

    MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification. (arXiv:2110.14795v1 [https://t.co/1uAOVrWCaz]) https://t.co/75FMKAEWTc
    view full post

    October 29, 2021

  • arXiv reaDer bot (cs-CV)
    @arXiv_reaDer (Twitter)

    MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification MedMNIST v2：2Dおよび3D生物医学画像分類のための大規模軽量ベンチマーク 2021-10-27T22:02:04+00:00 arXiv: https://t.co/ZVZXua1Rw6 英/日サマリ↓ https://t.co/794r0Cl1MS
    view full post

    October 28, 2021

Abstract Synopsis

  • MedMNIST v2 is a large collection of standardized biomedical images, including 12 datasets for 2D images and 6 for 3D images, all preprocessed into small, manageable sizes to make it easy for users to work with without needing background knowledge.
  • The dataset contains over 700,000 2D images and nearly 10,000 3D images, supporting various research and educational tasks like classification, with different complexities such as binary, multi-class, and multilabel problems.
  • The creators tested different methods, including neural networks and AutoML tools, to establish baseline performance, and all data and code are openly available for further research in biomedical image analysis and machine learning.]