Can Artificial Intelligence Guided Image Assessment be as Accurate as Laser Doppler Perfusion Scanning in Predicting Depth of Burn Injury?
Bibliographic record
Abstract
PURPOSE: Appropriate identification of burn depth and size is paramount. Despite the development of assessment aids [e.g., laser doppler imaging (LDI)], clinical assessments remain the gold standard, which assesses partial thickness burn depth with ~67% accuracy. We sought to develop an image-based artificial intelligence (AI) system that predicts burn severity and margins for use in acute burn triage. METHOD: A convoluted neural network (CNN) was trained on 1855 mobile-device-captured burn images of different burn depths. The CNN was used to develop a novel Boundary-Attention Mapping (BAM) algorithm, using elements of saliency mapping, which was utilized to recognize the boundaries of burns. For validation, 144 patient charts that included clinical assessments, burn location, total body surface area, LDI-assessments, were retrieved for a retrospective study at the University of Alberta. The clinical images underwent CNN-BAM assessments and were directly compared with the LDI assessment. RESULTS: The CNN-BAM system can highlight burns from surrounding tissue with high confidence. The CNN can classify four levels of burn severity with an accuracy of 80%. Results comparing the CNN-BAM outputs to clinical and LDI assessments have shown a high degree of correlation (approximately 85%) between the CNN burn severity predictions to those extrapolated from LDI healing potential. When compared to pre-LDI clinical assessment, the accuracy of the CNN-BAM outcomes has been equivalent or superior in most cases. A high degree of correlation has been demonstrated between the LDI scans and BAM maps created by the system when identifying the overall burn injury margins. CONCLUSION: This novel AI algorithm gives approximately equal accuracy in detecting burn depth as an LDI with a more economical and accessible application when embedded in a mobile device.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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 teacher head, 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".