A Comparative Study of Classical and Quantum Support Vector Machines for Multimodal Oral Cancer Detection
Bibliographic record
Abstract
Oral cancer, a persistent global health challenge ranking among the top ten cancers worldwide, requires accurate and robust early detection for improved patient prognosis. This study presents a preliminary feasibility investigation comparing classical Support Vector Machine (SVM) and Quantum Support Vector Machine (QSVM) for multimodal oral cancer detection, integrating photographic oral cavity image features extracted using ResNet-50 and clinical data features including age, gender, and risk factors (smoking, alcohol, betel chewing) with polygonal lesion annotations and patient metadata supporting the multimodal approach. The analysis reveals that SVM demonstrates adaptability, with performance metrics like accuracy, F1 score, precision, and recall varying across experiments (e.g., F1-scores range from 0.912 to 0.924), and handles class imbalance reasonably well, though occasional misclassification of minority class samples occurs. In contrast, QSVM exhibits minimal variability across experiments, with performance largely unaffected by feature transformations or dataset imbalance. This invariance may result from limitations of the quantum kernel in capturing imbalanced data complexities. Unlike prior studies that focus on either imaging or clinical data, this work explores the integration of multimodal features in a quantum-enhanced learning framework, representing an early exploration to apply QSVM for oral cancer detection. While QSVM shows theoretical promise, its practical utility is constrained in this context due to simulation-only experiments, limited qubit configurations, and dataset-specific challenges. Overall, the study highlights SVM's robustness for real-world imbalanced medical datasets while positioning QSVM as a promising direction for future quantumbased healthcare research.
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 | 0.004 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".