Summary of Biomarker Based Cancer Classification Using An Ensemble with Pre-trained Models, by Chongmin Lee and Jihie Kim
Biomarker based Cancer Classification using an Ensemble with Pre-trained Models
by Chongmin Lee, Jihie Kim
First submitted to arxiv on: 14 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
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 focuses on improving the detection of certain cancer types, such as pancreatic cancer, by identifying causal relationships between biomarkers and cancer. Liquid biopsies are a promising non-invasive method for monitoring specific biomarkers, enabling more precise medical interventions. To achieve this, the authors utilize machine learning algorithms like Random Forest and SVM for classification, but acknowledge that hyperparameter tuning is inefficient. They propose a meta-trained Hyperfast model for classifying cancer, achieving an AUC of 0.9929 on imbalanced datasets, outperforming other ML algorithms in binary classification tasks. Additionally, they introduce a novel ensemble model combining pre-trained Hyperfast, XGBoost, and LightGBM for multi-class classification, demonstrating improved accuracy (0.9464) using only 500 PCA features. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper tries to make it easier to find cancer early on by understanding how certain biomarkers are related to cancer. Biomarkers are like special signs in your body that can show if you have cancer or not. Right now, doctors often need to take a biopsy (a small sample of tissue) from the patient’s body to look for these signs. This can be tricky and may not always work well. The researchers found a way to use machine learning algorithms to identify biomarkers without needing a biopsy. They tested their method on some datasets and got good results, especially when dealing with imbalanced data (where there are more instances of one class than the other). |
Keywords
» Artificial intelligence » Auc » Classification » Ensemble model » Hyperparameter » Machine learning » Pca » Random forest » Xgboost
