Objective: To construct a prediction model based on the Padua Prediction Scale(PPS) for elderly hemiplegic patients with acute ischaemic stroke(AIS) complicated by lower limb deep vein thrombosis(DVT) and to verify its applicability. Methods: Clinical data from 220 elderly AIS hemiplegic patients who visited the Suzhou Integrated Traditional Chinese and Western Medicine Hospital from Jan 2020 to Dec 2023 were collected. The data was divided into a training set(176 cases) and a validation set(44 cases) in an 8∶2 ratio. The training set was further divided into DVT group and non-DVT group based on whether patients developed lower limb DVT. Univariate and multivariate Logistic regressions were performed using the PPS indicators, and a column-line graph risk prediction model was constructed using R software. Receiver operating characteristic(ROC) curve, calibration curve, and decision curves were evaluated using data from the validation set. Results: A total of 176 elderly AIS hemiplegic patients among which 48 cases complicated by lower limb DVT were included in the training set, and the incidence of DVT was 27.27%. Univariate analysis showed that the proportion of bed rest time>3 d, history of DVT, concomitant thrombophilic disease, cardiac or respiratory failure, acute infectious/rheumatic disease,complete hemiplegia,conscious disturbance and coma, and use of dehydrating drugs in DVT group were higher than those in non-DVT group, and the difference was statistically significant(P<0.05). Gender, age, body mass index(BMI) and history of hormone therapy were not statistically significant in the comparison between the two groups(P＞0.05). The degree of hemiparesis, use of dehydration drugs, level of consciousness, bed rest time, and cardiac or respiratory failure were entered into R software to construct a column-line graph, in which the AUC of the training set of elderly hemiplegic patients with AIS complicated by DVT was 0.96(0.93-0.99), the accuracy was 0.91(0.86-0.95), the sensitivity was 0.89(0.84-0.94) and the specificity was 0.96(0.90-0.99), and the cut-off value was 0.321. The slopes of the calibration curves of the column plots of the training set and validation set concurrent DVT were 0.999 and 0.678, which were close to 1. The results of the goodness-of-fit test showed that the difference between the predicted probability and the actual probability was not statistically significant when comparing the predicted probability and the actual probability(χ²=4.767,P=0.782). When the high-risk threshold was>0.10, those at risk of concomitant DVT could use this prediction model to better achieve clinical benefit. Conclusion: The prediction model of complicated by lower limb DVT in elderly hemiplegic patients with AIS constructed based on PPS has high clinical application value and can provide reference for clinical DVT evaluation and prevention and control.