Cleaning Up the MESS: Can Machine Learning Be Used to Predict Lower Extremity Amputation after Trauma-Associated Arterial Injury?
Bibliographic record
Abstract
BACKGROUND: Thirty years after the Mangled Extremity Severity Score was developed, advances in vascular, trauma, and orthopaedic surgery have rendered the sensitivity of this score obsolete. A significant number of patients receive amputation during subsequent admissions, which are often missed in the analysis of amputation at the index admission. We aimed to identify risk factors for and predict amputation on initial admission or within 30 days of discharge (peritraumatic amputation [PTA]). STUDY DESIGN: The Nationwide Readmission Database for 2016 and 2017 was used in our analysis. Factors associated with PTA were identified. We used XGBoost, random forest, and logistic regression methods to develop a framework for machine learning-based prediction models for PTA. RESULTS: We identified 1,098 adult patients with traumatic lower extremity fracture and arterial injuries; 206 underwent amputation. One hundred and seventy-six patients (85.4%) underwent amputation during the index admission and 30 (14.6%) underwent amputation within a 30-day readmission period. After identifying factors associated with PTA, we constructed machine learning models based on random forest, XGBoost, and logistic regression to predict PTA. We discovered that logistic regression had the most robust predictive ability, with an accuracy of 0.88, sensitivity of 0.47, and specificity of 0.98. We then built on the logistic regression by the NearMiss algorithm, increasing sensitivity to 0.71, but decreasing accuracy to 0.74 and specificity to 0.75. CONCLUSIONS: Machine learning-based prediction models combined with sampling algorithms (such as the NearMiss algorithm in this study), can help identify patients with traumatic arterial injuries at high risk for amputation and guide targeted intervention in the modern age of vascular surgery.
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 | 0.006 | 0.032 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| 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".