Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0631
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-08-03T06:08:36.049449Z
status: published
lens: trace
sector: health
headline: Exploratory machine learning-based early post-treatment assessment of willingness to reuse rubber dam isolation after microscopic root canal treatment
dek: ObjectiveTo explore factors associated with willingness to reuse rubber dam isolation after microscopic root canal treatment and develop an exploratory machine learning-based early post-treatment assessment model.MethodsThis retrospective cross-sectional study included 306 patients who underwent microscopic root canal treatment with rubber dam isolation from May 2025 to November 2025. The outcome was the willingness to reuse rubber dam isolation at the 1-week follow-up. Forty-seven newly enrolled patients were r…
gain_title: An exploratory Light Gradient Boosting Machine model predicted 1-week willingness to reuse rubber dam isolation after microscopic root canal treatment with AUC 0.939 in held-out test and 0.983 in temporal validation.
problem_title: Model performance dropped when satisfaction was excluded and estimates from the small same-center temporal validation cohort with only 14 unwilling patients were considered preliminary and potentially imprecise.
trace_subject: predicting willingness to reuse rubber dam isolation after microscopic root canal treatment using machine learning
gain_reading: An exploratory Light Gradient Boosting Machine model predicted 1-week willingness to reuse rubber dam isolation after microscopic root canal treatment with AUC 0.939 in held-out test and 0.983 in temporal validation.
gain_evidence: The Light Gradient Boosting Machine model showed the best exploratory performance, with area under the receiver operating characteristic curve values of 0.939 and 0.983 and F1-scores of 0.94 and 0.96 in the held-out test set and same-center temporal validation cohort, respectively.
problem_reading: Model performance dropped when satisfaction was excluded and estimates from the small same-center temporal validation cohort with only 14 unwilling patients were considered preliminary and potentially imprecise.
problem_evidence: Given the small temporal cohort, especially the 14 patients unwilling to reuse rubber dam isolation, this estimate was considered preliminary and potentially imprecise. | After satisfaction level was excluded, the area under the receiver operating characteristic curve decreased to 0.820 and 0.840 in the two evaluation cohorts.
quick_read: Researchers retrospectively analyzed 306 patients who had microscopic root canal treatment with rubber dam isolation between May and November 2025, defining willingness to reuse at 1-week follow-up as the outcome, with 246 willing and 60 unwilling. They trained six models on 26 variables and found the LightGBM model retained 12 predictors and achieved the highest exploratory AUCs of 0.939 and 0.983.

High early accuracy suggests satisfaction, postoperative pain, and discomfort-related factors could be flagged soon after treatment to improve patient counseling, but the result rests on a single-center design and a 47-patient temporal cohort, so generalizability and clinical utility remain unproven without larger prospective multicenter testing.
limitation: Temporal validation was based on only 47 patients including 14 unwilling to reuse, making performance estimates preliminary and imprecise, with prospective multicenter validation still required.
tag: Automated dual reading
key_points: Retrospective cross-sectional study of 306 patients from May 2025 to November 2025, with 259 in development cohort and 47 reserved as same-center temporal validation cohort. | 26 demographic, clinical, procedural, nursing-related, and post-treatment experience variables collected; recursive feature elimination retained 12 predictors. | SHAP analysis suggested satisfaction level and postoperative visual analog scale score were leading model-associated factors.
rundown: The study used 259 patients for development with a 7:3 training to held-out test split and 47 patients as a same-center temporal validation cohort from the same May to November 2025 enrollment window.

Preprocessing, recursive feature elimination with cross-validation, and hyperparameter tuning were integrated within a fully nested cross-validation pipeline using training data only, and six machine-learning models were compared with SHAP and generalized additive models for interpretation.
sources:
- peer_reviewed | Journal of International Medical Research | https://doi.org/10.1177/03000605261473155 | 2026-08-01
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

eba1b0a06046e03f92e70f103f29b4673ccee2d6a8284d71e1970ffd8023b169

previous

0000000000000000000000000000000000000000000000000000000000000000