Summary of An Autoencoder and Generative Adversarial Networks Approach For Multi-omics Data Imbalanced Class Handling and Classification, by Ibrahim Al-hurani et al.
An Autoencoder and Generative Adversarial Networks Approach for Multi-Omics Data Imbalanced Class Handling and Classification
by Ibrahim Al-Hurani, Abedalrhman Alkhateeb, Salama Ikki
First submitted to arxiv on: 16 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Neural and Evolutionary Computing (cs.NE); Genomics (q-bio.GN)
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 study integrates machine learning methodologies to tackle challenges in medical diagnostics, specifically dealing with high-dimensional data from multi-omics sequencing and imbalanced datasets. The research focuses on dimensionality reduction using autoencoder-based neural networks and Generative Adversarial Networks (GAN) to generate synthetic samples. The model starts by selecting discriminative features before feeding them into the network, aiming to predict cancer outcomes for different datasets. Notably, the proposed model outperforms existing models with accuracy scores of 95.09% for bladder cancer and 88.82% for breast cancer. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The study uses machine learning to improve medical diagnostics. It deals with a big problem in medicine: making sense of lots of data from DNA tests. This data is like a huge puzzle, and old methods can’t solve it well. The researchers used special techniques called autoencoders and GANs to make the puzzle smaller and more understandable. They also developed a way to add new “fake” samples to the data to make it more balanced. The goal was to create a better model that could predict cancer outcomes from DNA tests. And, surprisingly, this model did really well! It correctly predicted bladder cancer 95% of the time and breast cancer 89% of the time. |
Keywords
» Artificial intelligence » Autoencoder » Dimensionality reduction » Gan » Machine learning
