Detection of Low Blood Hemoglobin Levels on Pulmonary CT Angiography: A Feasibility Study Combining Dual-Energy CT and Machine Learning

Objectives: To evaluate if dual-energy CT (DECT) pulmonary angiography (CTPA) can detect anemia with the aid of machine learning. Methods: Inclusion 100 patients (mean age ± SD, 51.3 14.8 years; male-to-female ratio, 42/58) who underwent DECT CTPA and hemoglobin (Hb) analysis within 24 h, including 50 cases Hb below controls ≥ 12 g/dL. Blood pool attenuation was assessed on virtual noncontrast (VNC) images at eight locations. A classification model using extreme gradient-boosted trees developed a training set (n = 76) for differentiating from controls. The best evaluated in separate test 24). Results: significantly lower than (p-values < 0.01), except right atrium (p 0.06). learning had sensitivity, specificity, accuracy 83%, 92%, 88%, respectively. Measurements descending aorta highest relative importance among all features; threshold 43 HU yielded 68%, 76%, 72%, Conclusion: VNC imaging shows good diagnostic performance detecting CTPA.