Summary of Semi-supervised Multimodal Multi-instance Learning For Aortic Stenosis Diagnosis, by Zhe Huang et al.
Semi-Supervised Multimodal Multi-Instance Learning for Aortic Stenosis Diagnosis
by Zhe Huang, Xiaowei Yu, Benjamin S. Wessler, Michael C. Hughes
First submitted to arxiv on: 9 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Emerging Technologies (cs.ET); Machine Learning (cs.LG)
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 The proposed Semi-supervised Multimodal Multiple-Instance Learning (SMMIL) framework is a deep learning pipeline designed for the automatic interpretation of ultrasound imaging of the heart, specifically for detecting and treating aortic stenosis. This approach addresses two key limitations in existing methods: reliance on 2D cineloops and lack of utilization of Doppler imaging, which provides valuable information about pressure gradients and blood flow abnormalities associated with AS. The SMMIL framework combines spectral Dopplers and 2D cineloops as input modalities to produce a study-level AS diagnosis. During training, it leverages both labeled and unlabeled datasets to improve its performance. Experimental results demonstrate that SMMIL outperforms recent alternatives in 3-level AS severity classification and clinically relevant AS detection tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Semi-supervised Multimodal Multiple-Instance Learning is a new way to use computers to help doctors diagnose heart problems from ultrasound images. Right now, doctors rely too much on just looking at pictures of the heart taken with ultrasound machines. But these machines can also take special tests that show how blood flows through the heart, which is important for diagnosing some heart problems. The SMMIL method combines this extra information with the pictures to make a more accurate diagnosis. It’s like having a second pair of eyes to help doctors detect serious heart conditions earlier and more accurately. |
Keywords
* Artificial intelligence * Classification * Deep learning * Semi supervised
