Objective: To compare and analyze the differences in computed tomography (CT) imaging findings between pulmonary disease caused by the Mycobacterium avium-intracellulare complex (MAC) and pulmonary disease caused by other non-tuberculous mycobacterial (NTM) species, to explore the CT imaging characteristics of MAC pulmonary disease, and to improve the CT diagnostic accuracy for MAC pulmonary disease. Methods: With the retrospective study, the clinical data of 259 patients with confirmed NTM pulmonary diseases admitted to Beijing Chest Hospital, Capital Medical University from January 1, 2013 to August 31, 2021 were collected based on the inclusion criteria. Among them, 138 patients had MAC pulmonary disease (MAC group), and 121 patients had pulmonary disease caused by other NTM species (other NTM group). Univariate and multivariate binary logistic regression models were used to analyze the imaging features (including patchy opacities, ground-glass opacities, interstitial changes, thin-walled cavities, thick-walled cavities, mixed cavities, localized adjacent pleural thickening, bronchial wall thickening, extensive pleural thickening, emphysema, etc.) of the two groups. The diagnostic performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), and the DeLong test was employed to compare the differences in AUC values between the combined prediction and individual imaging signs. Results: The median age (interquartile range) of the MAC group was 61.00 (54.75, 68.25) years, which was higher than that of the other NTM group (52.00 (42.50, 61.00) years), with a statistically significant difference (Z=5.246,P<0.001). The proportions of CT imaging findings such as patchy opacities, interstitial changes, thick-walled cavities, and mixed cavities in the MAC group (97.10% (134/138), 19.57% (27/138), 65.22% (90/138), and 44.20% (61/138), respectively) were significantly higher than those in the other NTM group (89.26% (108/121), 4.96% (6/121), 35.54% (43/121), and 31.40% (38/121), respectively), with statistically significant differences (χ2=6.470, P=0.011; χ2=12.372, P<0.001; χ2=22.734, P<0.001; χ2=4.472, P=0.034). Multivariate binary logistic regression analysis showed that interstitial changes and thick-walled cavities were positively correlated with MAC pulmonary disease (β=1.045, OR (95%CI)=2.843 (1.023-7.900); β=2.239, OR (95%CI)=9.381 (2.845-30.937)), while mixed cavities were negatively correlated with MAC pulmonary disease (β=-1.452, OR (95%CI)=0.234 (0.076-0.721)). ROC curve analysis revealed that the AUC values for predicting MAC pulmonary disease using patchy opacities, interstitial changes, thick-walled cavities, and mixed cavities individually, as well as the combined detection of these four imaging signs, were 0.539 (95%CI: 0.469-0.610), 0.573 (95%CI: 0.504-0.642), 0.648 (95%CI: 0.581-0.716), 0.564 (95%CI: 0.494-0.634), and 0.705 (95%CI: 0.643-0.768), respectively. The DeLong test showed that the diagnostic performance of thick-walled cavities for MAC pulmonary disease was second only to that of the combined detection (Z=-3.117, P=0.002) and was significantly higher than that of patchy opacities, interstitial changes, and mixed cavities (Z=-3.337, P=0.001; Z=-2.296, P=0.022; Z=4.300, P<0.001). Conclusion: The combined model of patchy opacities, interstitial changes, thick-walled cavities, and mixed cavities demonstrates certain diagnostic efficacy in predicting MAC pulmonary disease, with thick-walled cavities being the imaging finding with the highest diagnostic value among them.