An Improved Framework for Detecting Thyroid Disease Using Filter-Based Feature Selection and Stacking Ensemble
Bibliographic record
Abstract
In recent years, machine learning (ML) has become a pivotal tool for predicting and diagnosing thyroid disease. While many studies have explored the use of individual ML models for thyroid disease detection, the accuracy and robustness of these single-model approaches are often constrained by data imbalance and inherent model biases. This study introduces a filter-based feature selection and stacking-based ensemble ML framework, tailored specifically for thyroid disease detection. This framework capitalizes on the collective strengths of multiple base models by aggregating their predictions, aiming to surpass the predictive performance of individual models. Such an approach can also reduce screening time and costs considering few clinical attributes are used for diagnosis. Through extensive experiments conducted on a clinical thyroid disease dataset, the filter-based feature selection approach and the ensemble learning method demonstrated superior discriminative ability, reflected by improved receiver operating characteristic-area under the curve (ROC-AUC) scores of 99.9%. The proposed framework sheds light on the complementary strengths of different base models, fostering a deeper understanding of their joint predictive performance. Our findings underscore the potential of ensemble strategies to significantly improve the efficacy of ML-based detection of thyroid diseases, marking a shift from reliance on single models to more robust, collective approaches.
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 | 0.002 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 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) | 0.001 | 0.001 |
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".