Alanine Aminotransferase and Bilirubin Dynamic Evolution Pattern as a Novel Model for the Prediction of Acute Liver Failure in Drug-Induced Liver Injury
Bibliographic record
Abstract
Aims: To develop, optimize, and validate a novel model using alanine aminotransferase (ALT) and total bilirubin (TB) dynamic evolution patterns in predicting acute liver failure (ALF) in drug-induced liver injury (DILI) patients. Methods: The demographics, clinical data, liver biopsy, and outcomes of DILI patients were collected from two hospitals. According to the dynamic evolution of ALT and TB after DILI onset, the enrolled patients were divided into ALT-mono-peak, TB-mono-peak, double-overlap-peak, and double-separate-peak (DSP) patterns and compared. Logistic regression was used to develop this predictive model in both discovery and validation cohorts. Results: The proportion of ALF was significantly higher in patients with the DSP pattern than in the ALT-mono-peak pattern and DOP pattern (10.0 vs. 0.0% vs. 1.8%, p < 0.05). The area under receiver operating characteristic curve (AUROC) of the DSP pattern model was 0.720 (95% CI: 0.682–0.756) in the discovery cohort and 0.828 (95% CI: 0.788–0.864) in the validation cohort in predicting ALF, being further improved by combining with international normalized ratio (INR) and alkaline phosphatase (ALP) (AUROC in the discovery cohort: 0.899; validation cohort: 0.958). Histopathologically, patients with the DSP pattern exhibited a predominantly cholestatic hepatitis pattern (75.0%, p < 0.05) with a higher degree of necrosis (29.2%, p = 0.084). Conclusion: DILI patients with the DSP pattern are more likely to progress to ALF. The predictive potency of the model for ALF can be improved by incorporating INR and ALP. This novel model allows for better identification of high-risk DILI patients, enabling timely measures to be instituted for better outcome.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.002 |
| 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".