Objective Little information is available about deep learning methods used in ultrasound images of salivary gland tumors. We aimed to compare the accuracy of the ultrasound‐trained model to computed tomography or magnetic resonance imaging trained model. Materials and methods Six hundred and thirty‐eight patients were included in this retrospective study. There were 558 benign and 80 malignant salivary gland tumors. A total of 500 images (250 benign and 250 malignant) were acquired in the training and validation set, then 62 images (31 benign and 31 malignant) in the test set. Both machine learning and deep learning were used in our model. Results The test accuracy, sensitivity, and specificity of our final model were 93.5%, 100%, and 87%, respectively. There were no over fitting in our model as the validation accuracy was similar with the test accuracy. Conclusions The sensitivity and specificity were comparable with current MRI and CT images using artificial intelligence.

Free flap reconstruction has been the mainstay among reconstruction surgeries for head and neck cancer. Intraoperative and postoperative hemoglobin (Hb) levels were both possible risk factors of flap failure and had been discussed widely. However, few investigations of preoperative Hb were seen in the previous study with its effect to flap condition remain uncertain and no conclusions in the literature. Patients who underwent free flap reconstruction after head and neck surgery in our institution between May 2014 and May 2019 were enrolled. The postoperative flap condition was observed carefully, and re-exploration was performed if necessary. We then retrospectively collected patient data with several intraoperative and postoperative indices. A total of 598 patients were enrolled in our study. The total major flap complication rate was 10.6%, with an overall success rate of 89.4%. They were predominantly male (95%), and most of them underwent free flap reconstruction for the first time (91%). A total of 81 (13%) patients received radiotherapy before reconstruction. Among all factors, the preoperative Hb level and free flap type showed significance in univariate and multivariate analyses. A previous history of radiotherapy, body mass index, nutrition status, or poorly controlled diabetes mellitus showed no significant results in either univariate or multivariate analysis. Our study showed that a lower preoperative Hb level affects free flap survival. Meanwhile, preoperative radiotherapy history has no significant influence in either univariate or multivariate analysis.