Objective:To explore the clinical application value of artificial intelligence models in identifying bleeding events and hemorrhagic points during laparoscopic pancreatic surgery.Methods:This single-center retrospective cohort study collected surgical videos of 25 patients undergoing laparoscopic pancreatic surgery at the Department of General Surgery, Peking Union Medical College Hospital from January 2022 to December 2024. Videos within 5 seconds before and after representative bleeding events were captured at 30 frames/s, with 11 666 hemorrhagic-related video frames annotated. Two algorithm models were developed: a pigment-based model and a pigment+optical flow-based model for classification and target recognition of bleeding frames. The training and test sets for the pigment-based algorithm contained 4 692 hemorrhagic and 4 309 non-hemorrhagic frames, while those for the pigment+optical flow model included 1 339 hemorrhagic and 1 326 non-hemorrhagic frames. Performance evaluation was conducted using overlap thresholds, with accuracy and recall rates as key metrics.Results:The pigment-based model achieved 93.8% accuracy (134/143) and 43.3% recall (134/310) in hemorrhagic frame classification. At an overlap threshold of 0.3, the pigment-based model showed 84.1% accuracy (433/515) and 85.4% recall (433/507) in target recognition. When the threshold was increased to 0.5, the pigment+optical flow model demonstrated 88.1% accuracy (354/402) and 89.2% recall (354/397) in hemorrhagic target recognition.Conclusions:It is difficult to distinguish active bleeding from old bleeding completely by pigment information alone. The spatio-temporal features can be effectively extracted by combining pigment and optical flow information, and the bleeding can be accurately identified and located, which has potential clinical application value.

Objective:To explore the clinical application value of artificial intelligence models in identifying bleeding events and hemorrhagic points during laparoscopic pancreatic surgery.Methods:This single-center retrospective cohort study collected surgical videos of 25 patients undergoing laparoscopic pancreatic surgery at the Department of General Surgery, Peking Union Medical College Hospital from January 2022 to December 2024. Videos within 5 seconds before and after representative bleeding events were captured at 30 frames/s, with 11 666 hemorrhagic-related video frames annotated. Two algorithm models were developed: a pigment-based model and a pigment+optical flow-based model for classification and target recognition of bleeding frames. The training and test sets for the pigment-based algorithm contained 4 692 hemorrhagic and 4 309 non-hemorrhagic frames, while those for the pigment+optical flow model included 1 339 hemorrhagic and 1 326 non-hemorrhagic frames. Performance evaluation was conducted using overlap thresholds, with accuracy and recall rates as key metrics.Results:The pigment-based model achieved 93.8% accuracy (134/143) and 43.3% recall (134/310) in hemorrhagic frame classification. At an overlap threshold of 0.3, the pigment-based model showed 84.1% accuracy (433/515) and 85.4% recall (433/507) in target recognition. When the threshold was increased to 0.5, the pigment+optical flow model demonstrated 88.1% accuracy (354/402) and 89.2% recall (354/397) in hemorrhagic target recognition.Conclusions:It is difficult to distinguish active bleeding from old bleeding completely by pigment information alone. The spatio-temporal features can be effectively extracted by combining pigment and optical flow information, and the bleeding can be accurately identified and located, which has potential clinical application value.

Ultra-performance liquid chromatography-quadrupole-electrostatic field Orbitrap mass spectrometry(UHPLC-Q-Exactive Orbitrap MS/MS) was used for rapid identification of the chemical components in Kaixin San substance benchmark. The gradient elution was performed through a Waters ACQUITY~(TM) BEH C_(18) column(2.1 mm×150 mm, 1.7 μm) with water-acetonitrile as mobile phase, a column temperature of 30 ℃, a flow rate of 0.3 mL·min~(-1), and a sample size of 1 μL. The scanning was performed in the negative ion mode. The complex component groups in Kaixin San substance benchmark were quickly and accurately identified and clearly assigned based on the comparison of the retention time and MS data with those of the reference substance as well as the relative molecular weight of the same or similar components in the mass spectrum database and literature. A total of 77 compounds were identified, including 26 saponins, 13 triterpenoid acids, 20 oligosaccharide esters, 5 xanthones, and 13 other compounds. The qualitative method established in this study can systematically, accurately, and quickly identify the chemical components in Kaixin San substance benchmark, which can provide a basis for the further analysis of its active components in vivo and the establishment of its quality control system.
Benchmarking ; Chromatography, High Pressure Liquid/methods* ; Drugs, Chinese Herbal/chemistry* ; Tandem Mass Spectrometry/methods*