Yang et al. analyzed the satisfaction of patients a year after distal radius fracture surgery. In order to do this, Yang et al. utilized a multimodal machine learning model that utilized postoperative X-ray radiomics with rating scores given to them from the patients themselves. Radiomics is a specific method that can convert medical imaging like X-rays, CT scans, or MRIs into quantitative data to be analyzed by a computer or in this study, a machine learning model. The patient rating scores were determined on the basis of measured wrist pain throughout daily life, general comfortability of daily life, and how often the patient would forget about the wrist pain throughout daily life (the joint would feel more natural and less painful if patients were to forget about it more). The study consisted of 385 patients who were split into two groups categorized by internal fixation and external fixation. 169 of these patients had follow up assessments and further imaging of the distal radius fracture. From the X-rays, 848 radiomic features were specifically found and categorized into 16 key features with random forest embedding. These 16 key features were created in order to train and test the functionality of the machine learning models, which used decision curve analysis, calibration curves, and ROC curves to evaluate performance across the 16 key features. The results showed that the model that involved radiomics in addition to patient functional rating scales actually surpassed the model with only radiomics in its evaluations in terms of predictive reliability and efficiency in a clinical setting. This showed that the combination of radiomics imaging data with patient functional rating scales was a stronger predictor of long term patient satisfaction specifically after the distal radius fracture.