Leveraging Machine Learning for Severity Level-Wise Biomarker Identification in Prostate Cancer Microarray Gene Expression Data
Bibliographic record
Abstract
Background: Prostate cancer is the most commonly occurring cancer amongst men. The detection and treatment of this cancer is therefore of great importance. The severity level of this cancer, which is established as a score in the Gleason Grading Group (GGC), guides the treatment of the cancer. Methods: In this paper, traditional machine learning (ML) classification methods such as Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and XGBoost (XGB), which have recently been shown to accurately identifying biomarkers for computational biology, are leveraged to find potential biomarkers for the different GGC scores. A ML framework that maps the Gleason Grading Group (GGG) into five severity levels—low, intermediate-low, intermediate, intermediate-high, and high—has been developed using the above methods. The microarray data for this ML method have been derived from immunohistochemical tests. The study includes severity level-wise biomarker identification, incorporating missing value imputation, class imbalance handling using the SMOTE-Tomek link method, and stratified k-fold validation to ensure robust biomarker selection. Results: The framework is evaluated on prostate cancer tissue microarray gene expression data from 1119 samples. A combination of high-aggressive and low-aggressive signatures are used in four experimental setups. The results demonstrate the effectiveness of the approach in distinguishing between critical biomarkers with highly accurate models, obtaining 96.85% accuracy using the XGBoost method. Conclusions: Leveraging ML gives a potential ground to involve the domain experts and the satisfactory results have approved that. For the future physician-in-the-loop approach can be tested to ensure further diagnosis impact.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".