The study developed and validated a three-dimensional fractal dimension (FD 3D) computed tomography (CT)-based nomogram for noninvasive prediction of tumor spread through airspaces (STAS) in 110 patients with stage IA lung adenocarcinoma after surgical resection. Patients were divided into STAS-positive group (48 patients) and STAS-negative group (62 patients) according to pathological examination. Univariate analysis showed significant differences in consolidation-to-tumor ratio (CTR, p < 0.001), morphological irregularity (p = 0.006), lobulation (p = 0.006), vascular convergence (p = 0.010), and FD 3D (p < 0.001) between groups. Multivariate analysis confirmed CTR, morphological irregularity, lobulation and FD 3D as independent predictors of STAS. The nomogram achieved an area under the curve (AUC) of 0.894 (95% CI: 0.821–0.944; p < 0.001) with a sensitivity of 75% and a specificity of 90.32%. At a cut-off value of 0.56, it had a positive predictive value of 85.71% in the high-risk group (42 patients, 38.18%) and a negative predictive value of 82.35% in the low-risk group (68 patients, 61.82%), with STAS prevalence of 85% versus 17.65% (χ² = 46.18, p < 0.001). The nomogram integrates FD 3D and CT features to predict STAS and may aid in surgical decision-making, but requires prospective studies.