Summary of Self-supervised Multiple Instance Learning For Acute Myeloid Leukemia Classification, by Salome Kazeminia et al.
Self-Supervised Multiple Instance Learning for Acute Myeloid Leukemia Classification
by Salome Kazeminia, Max Joosten, Dragan Bosnacki, Carsten Marr
First submitted to arxiv on: 8 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper explores the application of Self-Supervised Learning (SSL) to Multiple Instance Learning (MIL)-based Acute Myeloid Leukemia (AML) subtype classification from blood smears. The authors investigate three state-of-the-art SSL methods, SimCLR, SwAV, and DINO, and compare their performance against supervised pre-training. The findings show that SSL-pretrained encoders achieve comparable performance, demonstrating the potential of SSL in MIL-based disease diagnosis. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study uses machine learning to help doctors diagnose a type of blood cancer called AML. They’re trying to find a way to do this without needing lots of labeled pictures of AML cells. Normally, you need these labels to train your computer model, but it’s hard and expensive to get them. The researchers look at three ways to “pre-train” their model using fake tasks, like recognizing objects in pictures or matching similar-looking images. They compare how well these methods work against the usual way of training a model with labeled data. The results show that these new methods can be just as good at diagnosing AML without needing lots of labeled pictures. |
Keywords
» Artificial intelligence » Classification » Machine learning » Self supervised » Supervised
