S615 The Validation of CARS Score for Prediction of Patients With Achalasia
Bibliographic record
Abstract
Introduction: A novel endoscopic score system (CARS) was recently developed to predict achalasia based on grades of contents, anatomy, resistance, and stasis. We aimed to test the predictive power of CARS score in a large cohort of patients with esophageal symptoms. Methods: Adults undergoing upper endoscopy and HRM for an indication of esophageal symptoms were included. The endoscopic video from each patient was carefully reviewed by 2 general gastroenterologists, and any disagreements were confirmed by one esophagologist to determine the CARS score. HRM diagnoses by CCv4.0 were collected for each patient. Results: A total of 271 patients with no prior history of foregut surgery or esophageal interventions were included between August 2018 and July 2023. According to CCv4.0, patients were categorized as achalasia (n=89; Type 1=33, Type 2=46, Type 3=10), EGJOO (n=2), ineffective esophageal motility (IEM) (n=55), absent contractility (n=11), hypercontractile esophagus (n=5), distal esophageal spasm (DES) (n=7), and no motility disorder (n=102). Achalasia patients had a higher mean CARS score compared to those with no motility disorder (mean 4.5 vs 0.3, P = < 0.01). Absent contractility and IEM had a high CARS score compared to no motility disorder (mean 1.1 and 0.7 vs 0.3, P < 0.05). Sub-score analysis showed higher scores in Type 1 and Type 2 compared to Type 3 in terms of contents, anatomy and stasis (contents: mean 1.5 and 1.4 vs 0.9; anatomy: mean 1.2 and 1.0 vs 0.6; stasis: mean 0.9 and 0.8 vs 0.2, all P < 0.01). A composite CARS score of 4 or greater resulted in a sensitivity of 76.4%, specificity of 100%, positive predictive value (PPV) of 100%, negative predictive value (NPV) of 82.9% when comparing patients with achalasia and those with no motility disorder, and a sensitivity of 76.4%, specificity of 99.5%, PPV of 98.5%, NPV of 89.6% when comparing patients with achalasia and those with non-achalasia. Conclusion: The novel endoscopic CARS score provides high sensitivity and specificity in detecting achalasia. It has the potential to facilitate the timely and efficient diagnosis of achalasia and may reduce the necessity for manometry in certain scenarios (see Figure 1, Table 1).Figure 1.: Mean CARS score for each motility disorder by CC 4.0. * P < 0.05, ** P < 0.01. Table 1. - Areas under the receiver operating characteristic curve (AUROCs), sensitivity, and specificity, positive predictive value (PPV) and negative predictive value (NPV) for detecting achalasia CARS score AUROCs (95%CI) Cut-off value Sensitivity (%) Specificity (%) PPV (%) NPV (%) Achalasia vs No motility disorder 0.968 (0.942-0.994) 1.5 88.8 97.1 96.3 90.8 2.5 84.3 99 98.7 87.8 3.5 76.4 100 100 82.9 4.5 58.4 100 100 73.4 Achalasia vs Non-achalasia 0.955 (0.925-0.984) 1.5 88.8 91.8 84.0 94.4 2.5 84.3 96.7 92.6 92.6 3.5 76.4 99.5 98.5 89.6 4.5 58.4 99.5 98.1 83.0
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 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".