Fig. 5

Comparison of the accuracies of QCIGISH and current small biopsy pathology for lung cancer diagnosis. A Analysis showing improved or comparable sensitivities of QCIGISH over small biopsy pathology for different stages of NSCLC. B Analysis showing improved or comparable sensitivity of QCIGISH over small biopsy pathology for different stages of SCLC. C Analysis showing improved or comparable sensitivities of QCIGISH over small biopsy pathology for pulmonary nodules and masses. D Analysis showing comparable specificities of QCIGISH with small biopsy pathology for pulmonary nodules and masses. E List of clinically available minimally invasive sampling procedures for lung lesions. F An illustrated example of clinical cases both positively classified as lung cancer by QCIGISH but were diagnosed as benign by small biopsy pathology. Surgical histopathology or clinical diagnosis with 2-year follow-up were used as golden standard. CIS, carcinoma in situ. N.E.D., no enough data. Error bars on the bar charts showed the 95% CI. Scale bar, 50 μm