Authors: Ibrahim Al-Hurani, Abedalrhman Alkhateeb, Salama Ikki

Published on: May 16, 2024

Impact Score: 7.4

Arxiv code: Arxiv:2405.09756

Summary

• What is new: The integration of autoencoders with Generative Adversarial Networks (GANs) for handling high dimensional molecular biology data and class imbalance in medical diagnostics.
• Why this is important: High dimensionality and class imbalance in molecular biology datasets hinder effective diagnosis.
• What the research proposes: A neural network model that uses autoencoders to reduce data dimensionality and GANs to generate synthetic samples for balancing classes.
• Results: Achieved 95.09% accuracy on bladder cancer and 88.82% on breast cancer datasets, outperforming existing models.

Technical Details

Technological frameworks used: Neural Networks, Autoencoders, Generative Adversarial Networks

Models used: Feature selection models, predictive models for cancer outcomes

Data used: Multi-omics sequencing data from molecular biology

Potential Impact

Healthcare industry, particularly companies involved in medical diagnostics and pharmaceuticals, as well as providers of data analytics solutions

Want to implement this idea in a business?

We have generated a startup concept here: OncolytiX.