Deep Learning Algorithm for Abdominal Trauma Diagnosis on CT Imaging
Deep learning algorithms demonstrate high accuracy in detecting renal, liver, and spleen injuries on abdominal CT scans. With a notable performance in renal injury detection (accuracy: 0.932), the model exhibits potential for rapid preliminary screening and adjunctive diagnosis of traumatic abdominal injuries. The results suggest promising applicability in emergency settings, aiding clinicians in timely and accurate assessment of multiple organ injuries from CT imaging data.
Patients who needed emergency general surgery within 60 days of a pulmonary embolism had more major complications than those operated on later, in a national database study.
Among 3,785 patients who had emergency general surgery after acute pulmonary embolism, those operated on at least 60 days later had markedly fewer major adverse events (adjusted OR 0.45, 95% CI 0.37–0.55).
Appendix, Gallbladder and Surgical Emergencies3 Nov 2026
