Temporomandibular disorders (TMD), a common condition in oral and maxillofacial surgery, significantly impairs patients' quality of life. Early prediction and appropriate treatment of TMD are therefore critically important. Research on TMD prediction models has evolved from traditional statistical methods to machine learning and subsequently to deep learning, each offering unique advantages and limitations. Traditional statistical methods can effectively identify independent risk factors influencing treatment outcomes but generally rely on substantial prior knowledge and assumptions. Machine learning techniques can process large-scale, high-dimensional data and autonomously learning patterns and regularities within datasets. However, they exhibit strong dependence on data quality and limited model generalization capabilities. Deep learning approaches excel at automatically extracting temporal patterns and trends from time-series data while effectively capturing complex nonlinear relationships, yet they require extensive training datasets and suffer from interpretability challenges due to their inherent black-box testing. This review comprehensively evaluates the implementation and performance of these computational approaches in TMD prediction, critically analyzes their respective strengths and constraints, and discusses promising future research directions.

Objective:To evaluate the performance differences of three common novel quantitative ultrasound（QUS）imaging methods-Nakagami-Gamma parametric imaging，ultrasound attenuation intercept imaging，and weighted Shannon entropy imaging-in monitoring high-intensity focused ultrasound（HIFU）therapy.Methods:On a clinical HIFU therapy system， ex vivo pork loin tissue and in vivo rabbit leg tissue were treated with HIFU at different acoustic power levels（77-174 W）. Ultrasound monitoring radiofrequency（RF）data were acquired online，and the three QUS images were reconstructed based on identical raw data. Performance was compared through quantitative analysis of tissue damage contrast-to-noise ratio（CNR）and damage area assessment relative to true histological damage. Results:During HIFU therapy，all three QUS imaging methods outperformed B-mode ultrasound in characterizing HIFU-induced damage，with significantly higher CNR values. Among the images，Nakagami-Gamma images showed the highest increase in CNR values before and after treatment，with an improvement of 22.5%- 60.3%；meanwhile，the damage area assessed by weighted Shannon entropy images exhibited the highest correlation with the true damage area（ r=0.81， P<0.000 1）. Conclusions:By characterizing tissue microstructure features，QUS imaging can more reliably monitor HIFU thermal damage than conventional B-mode ultrasound. Among the methods，Nakagami imaging was the most sensitive to damage characterization，entropy parameter imaging showed the strongest correlation with the true damage area，and attenuation intercept imaging achieved the best damage area matching. This study may provide references for developing next-generation clinical HIFU therapy monitoring systems.

Temporomandibular disorders (TMD), a common condition in oral and maxillofacial surgery, significantly impairs patients' quality of life. Early prediction and appropriate treatment of TMD are therefore critically important. Research on TMD prediction models has evolved from traditional statistical methods to machine learning and subsequently to deep learning, each offering unique advantages and limitations. Traditional statistical methods can effectively identify independent risk factors influencing treatment outcomes but generally rely on substantial prior knowledge and assumptions. Machine learning techniques can process large-scale, high-dimensional data and autonomously learning patterns and regularities within datasets. However, they exhibit strong dependence on data quality and limited model generalization capabilities. Deep learning approaches excel at automatically extracting temporal patterns and trends from time-series data while effectively capturing complex nonlinear relationships, yet they require extensive training datasets and suffer from interpretability challenges due to their inherent black-box testing. This review comprehensively evaluates the implementation and performance of these computational approaches in TMD prediction, critically analyzes their respective strengths and constraints, and discusses promising future research directions.

Objective:To evaluate the performance differences of three common novel quantitative ultrasound（QUS）imaging methods-Nakagami-Gamma parametric imaging，ultrasound attenuation intercept imaging，and weighted Shannon entropy imaging-in monitoring high-intensity focused ultrasound（HIFU）therapy.Methods:On a clinical HIFU therapy system， ex vivo pork loin tissue and in vivo rabbit leg tissue were treated with HIFU at different acoustic power levels（77-174 W）. Ultrasound monitoring radiofrequency（RF）data were acquired online，and the three QUS images were reconstructed based on identical raw data. Performance was compared through quantitative analysis of tissue damage contrast-to-noise ratio（CNR）and damage area assessment relative to true histological damage. Results:During HIFU therapy，all three QUS imaging methods outperformed B-mode ultrasound in characterizing HIFU-induced damage，with significantly higher CNR values. Among the images，Nakagami-Gamma images showed the highest increase in CNR values before and after treatment，with an improvement of 22.5%- 60.3%；meanwhile，the damage area assessed by weighted Shannon entropy images exhibited the highest correlation with the true damage area（ r=0.81， P<0.000 1）. Conclusions:By characterizing tissue microstructure features，QUS imaging can more reliably monitor HIFU thermal damage than conventional B-mode ultrasound. Among the methods，Nakagami imaging was the most sensitive to damage characterization，entropy parameter imaging showed the strongest correlation with the true damage area，and attenuation intercept imaging achieved the best damage area matching. This study may provide references for developing next-generation clinical HIFU therapy monitoring systems.

Objective:To investigate the application of virtual reality (VR) technology on intraoperative pain in patients undergoing peritoneal dialysis (PD)-related procedures with local infiltration anesthesia and the satisfaction.Methods:It was a single-center, prospective, concurrent controlled study. Patients were divided into two groups: VR group and control group. In the VR group, patients wore a VR headset to watch soothing audio and video content during surgery, while the control group underwent routine procedures. Intraoperative pain and satisfaction were assessed using the visual analog scale (VAS) and a 5-point satisfaction scale within 30 minutes of surgery. In addition, tolerance of the VR experience in the VR group was assessed using the VR sickness questionnaire.Results:A total of 43 patients were included in the study, including 25 males (58.1%). Chronic glomerulonephritis [17 cases (39.5%)] and diabetic nephropathy [6 cases (14.0%)] were the main primary diseases. There were 23 cases in the control group and 20 cases in the VR group. There were no significant differences between the two groups in age, sex ratio, proportion of primary disease, diabetes, hypertension, distribution of operation methods, preoperative vital signs and operation time (all P>0.05). VAS pain score was significantly lower in the VR group than that in the control group (5.90±2.38 vs. 7.43±1.67, t=2.469, P=0.018). The percentage of patients who were satisfied was 89.5% (17/19) in the VR group and 78.3% (18/23) in the control group, but there was no significant difference (chi-square test for continuity correction, χ2=0.308, P=0.579). Three patients in the VR group withdrew from the study due to severe discomfort, while the remaining participants found the VR experience to be tolerable. Common adverse effects included fatigue and blurred vision. Conclusions:The application of VR technology in PD-related procedures has been effective in reducing intraoperative pain when combined with local infiltration anesthesia. Furthermore, the utilization of VR technology in PD-related procedures is associated with a safe and tolerable outcome, despite the observation of some adverse effects.