Specific spectral sub-images for machine learning evaluation of optical differences between carbon ion and X ray radiation effects
Bibliographic record
Abstract
Advances in radiotherapy, particularly the exploration of alternative radiation types such as carbon ions have updated our understanding of its effects and applicability on chondrosarcoma cells. Here we compare the optical effects produced by carbon ions (CI) and X-rays (XR) radiations on chondrosarcoma cells nuclei and set an automated method for evaluating the radiation-induced alterations without the need of chemical marking. Hyperspectral images (HSI) of SW1353 chondrosarcoma line carry detectable optical changes of the cells irradiated either with CI or XR compared to non-irradiated ones (REF). The differences between the spectral profiles of CI, XR and REF nuclei classes led to partitioning the HSIs into spectral sub-images. The changes are detected by support vector machine (SVM) classifiers whose performances are evaluated by the most used point metrics: sensitivity ( SEN ), accuracy ( ACC ), and precision ( PREC ), applied on spatial feature values. Specific interaction mechanisms by radiation type reveal distinct subintervals where HSIs changes are more prominent, and the classifiers perform at best. For CI the best classifiers are obtained for sub-images in the interval (424–436 nm), while for XR the best classifiers are obtained for sub-images in the interval (436–445 nm). The classifiers work better with texture features than roughness features in both cases. The classifier with the best SEN point metric in the testing phase is the most suitable to measure the irradiation efficiency irrespective of the radiation type. The altered nuclei are easier to discriminate when irradiated with CI than with XR. The study proves that SVM with optical data offers a rapid, automated, and label-free method for evaluating radiation-induced alterations in chondrosarcoma nuclei, thereby enabling effective analysis of extensive data.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".