Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1918
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-19T18:13:18.892568Z
status: published
lens: g_space
sector: lifestyle
headline: Three-dimensional anatomical classification and machine learning prediction of the P6a-right hepatic vein relationship: surgical implications for laparoscopic right anterior sectionectomy
dek: Background The spatial relationship between the ventral branch of segment VI portal vein (P6a) and the right hepatic vein (RHV) significantly influences the anatomical complexity of the right hepatic fissure and the surgical outcomes of laparoscopic right anterior sectionectomy (LRAS). The ventral-P6a variant is often associated with higher surgical risks, yet preoperative prediction remains challenging. Methods This retrospective study included 100 patients who underwent triple-phase enhanced CT of the upper ab…
gain_title: A logistic regression model using 3D CT anatomical measurements predicted whether the P6a portal branch lay ventral or dorsal to the right hepatic vein, offering a basis for preoperative risk assessment in laparoscopic right anterior sectionectomy.
problem_title: (none)
trace_subject: (none)
gain_reading: A logistic regression model using 3D CT anatomical measurements predicted whether the P6a portal branch lay ventral or dorsal to the right hepatic vein, offering a basis for preoperative risk assessment in laparoscopic right anterior sectionectomy.
gain_evidence: The logistic regression model demonstrated the best performance in the test set (F1-score = 0.741) with minimal overfitting. | providing a basis for preoperative risk assessment and surgical planning in LRAS
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers retrospectively analyzed triple-phase enhanced upper-abdominal CT scans from 100 patients and built 3D reconstructions to classify the P6a portal branch as dorsal or ventral relative to the right hepatic vein. They then trained five machine learning models on angular and linear anatomical measurements to predict that classification.

The ventral-P6a variant is described as carrying higher surgical risk, yet it is hard to predict before surgery, so an interpretable prediction tool could help surgeons plan. The evidence comes from a small retrospective sample, and the report does not show whether the tool changes surgical outcomes in practice.
limitation: The study was retrospective and included only 100 patients, and it assessed prediction of anatomical classification rather than demonstrating improved surgical outcomes.
tag: Evidence-backed gain
key_points: In the 100 patients, 59% were dorsal-P6a and 41% were ventral-P6a. | The ventral-P6a group had a smaller angle between the anterior and posterior portal branches (82.27° vs. 98.02°, P = 0.003). | A P6a origin less than 5 mm from the posterior branch origin was more common in the ventral group (37% vs. 10%, P = 0.001). | Logistic regression performed best in the test set, with an F1-score of 0.741.
rundown: The study compared five models: logistic regression, random forest, K-nearest neighbors, support vector machine, and XGBoost. Performance was evaluated with AUC and F1-score, and SHAP was applied to interpret which features drove predictions.

The most important predictive features were the angle between the anterior and posterior portal branches, the angle of the right hepatic fissure plane, and the diameter of P6a. The ventral group also had a larger angle between the right hepatic vein and the inferior vena cava (52.23° vs. 43.97°, P = 0.013).

The authors conclude that the interpretable model identifies key anatomical risk factors and could help optimize surgical strategies for high-risk patients with the ventral-P6a variant.
sources:
- peer_reviewed | Surgical and Radiologic Anatomy | https://doi.org/10.1007/s00276-026-04016-4 | 2026-10-14
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

659ab8edfb43616b2748dbbbfc2c92d4f8cd17b69363dbf50403a20d770a6d6d

previous

0000000000000000000000000000000000000000000000000000000000000000