Medical ultrasound technology is widely used for diagnosis and therapy in clinical practice.Ultrasound probes,which are directly contact with patients,pose a potential risk of pathogen transmission.This expert consen-sus was developed by a multidisciplinary team based on international guidelines,standards in China,and the results of a national survey,aiming to reduce the risk of healthcare-associated infection through standardizing reprocessing of medical ultrasound probes,and formulating consensus recommendations with the Delphi method.The consensus clarifies the reprocessing principles for three types of ultrasound probes of different infection risks:external-use ul-trasound probes,interventional percutaneous ultrasound probes,and internal-use ultrasound probes,puts forward systematic suggestions on the reprocessing standards and disinfection levels of ultrasound probe isolation covers and coupling agents,the reprocessing procedures and methods of ultrasound probes,as well as architectural layout and management of reprocessing,so as to provide a scientific prevention and control framework for ensuring ultrasound diagnosis and therapy safety.

Objective To propose a novel method based on three-dimensional depthwise separable convolution network(3D DSN)for the separation of PET images with dual tracers of 18F-FDG and 18F-FAPI.Methods A total of 120 pairs of 18F-FDG and 18F-FAPI PET images of the same patient scanned separately at different time points were collected,and the dual-tracer PET image was generated through simulation.After the image registration of PET images of two tracers for ensuring spatial position matching,the registered PET images were forward-projected to generate sinogram data,and the sinogram data of two tracers were accumulated to obtain mixed sinogram data.Subsequently,the dual-tracer PET image was reconstructed using maximum likelihood expectation maximization and input into a 3D DSN based network for image separation,thereby obtaining PET images of two single tracers.Results Compared with 3D CNN method,the proposed method increased the structure similarity index measure(SSIM)of the separated 18F-FDG images to the real 18F-FDG images by 0.87%,increased the peak signal-to-noise ratio(PSNR)by 11.8%,and reduced the normalized root mean square error(NRMSE)by 52%.The SSIM of the separated 18F-FAPI images to the real 18F-FAPI images increased by 1.1%,PSNR increased by 17.0%,and NRMSE decreased by 51%.Conclusion The proposed method can be effectively applied to simultaneous PET imaging with dual PET tracers,reducing the number of scans and costs in time and money,and providing clinical doctors more accurate and abundant diagnostic information.

Medical ultrasound technology is widely used for diagnosis and therapy in clinical practice.Ultrasound probes,which are directly contact with patients,pose a potential risk of pathogen transmission.This expert consen-sus was developed by a multidisciplinary team based on international guidelines,standards in China,and the results of a national survey,aiming to reduce the risk of healthcare-associated infection through standardizing reprocessing of medical ultrasound probes,and formulating consensus recommendations with the Delphi method.The consensus clarifies the reprocessing principles for three types of ultrasound probes of different infection risks:external-use ul-trasound probes,interventional percutaneous ultrasound probes,and internal-use ultrasound probes,puts forward systematic suggestions on the reprocessing standards and disinfection levels of ultrasound probe isolation covers and coupling agents,the reprocessing procedures and methods of ultrasound probes,as well as architectural layout and management of reprocessing,so as to provide a scientific prevention and control framework for ensuring ultrasound diagnosis and therapy safety.

Objective To propose a novel method based on three-dimensional depthwise separable convolution network(3D DSN)for the separation of PET images with dual tracers of 18F-FDG and 18F-FAPI.Methods A total of 120 pairs of 18F-FDG and 18F-FAPI PET images of the same patient scanned separately at different time points were collected,and the dual-tracer PET image was generated through simulation.After the image registration of PET images of two tracers for ensuring spatial position matching,the registered PET images were forward-projected to generate sinogram data,and the sinogram data of two tracers were accumulated to obtain mixed sinogram data.Subsequently,the dual-tracer PET image was reconstructed using maximum likelihood expectation maximization and input into a 3D DSN based network for image separation,thereby obtaining PET images of two single tracers.Results Compared with 3D CNN method,the proposed method increased the structure similarity index measure(SSIM)of the separated 18F-FDG images to the real 18F-FDG images by 0.87%,increased the peak signal-to-noise ratio(PSNR)by 11.8%,and reduced the normalized root mean square error(NRMSE)by 52%.The SSIM of the separated 18F-FAPI images to the real 18F-FAPI images increased by 1.1%,PSNR increased by 17.0%,and NRMSE decreased by 51%.Conclusion The proposed method can be effectively applied to simultaneous PET imaging with dual PET tracers,reducing the number of scans and costs in time and money,and providing clinical doctors more accurate and abundant diagnostic information.

Various types of software needed in radiomics studies come with the problems such as data incompatibility and hyperparameter tuning.Therefore,an artificial intelligence-based software is developed for radiomics studies,providing doctors and researchers a solution with image preprocessing,feature extraction,feature selection,modeling analysis and data visualization.The usability of the software is demonstrated using a public data set.Eight sets of feature selectors and classifiers are established for classification predication on test data set and key performance indicator output.Through hyperparameter tuning,the model is further optimized.Researchers will focus more on the research itself rather than unnecessary development efforts,and radiomics studies will become more convenient and efficient with the software addressed.

Objective To develop a method for accurate identification of multiscale carotid plaques in ultrasound images.Methods We proposed a two-stage carotid plaque detection method based on deep convolutional neural network(SM-YOLO).A series of algorithms such as median filtering,histogram equalization,and Gamma transformation were used to preprocess the dataset to improve image quality.In the first stage of the model construction,a candidate plaque set was built based on the YOLOX_l target detection network,using multiscale image training and multiscale image prediction strategies to accommodate carotid artery plaques of different shapes and sizes.In the second stage,the Histogram of Oriented Gradient(HOG)features and Local Binary Pattern(LBP)features were extracted and fused,and a Support Vector Machine(SVM)classifier was used to screen the candidate plaque set to obtain the final detection results.This model was compared quantitatively and visually with several target detection models(YOLOX_l,SSD,EfficientDet,YOLOV5_l,Faster R-CNN).Results SM-YOLO achieved a recall of 89.44%,an accuracy of 90.96%,a F1-Score of 90.19%,and an AP of 92.70%on the test set,outperforming other models in all performance indicators and visual effects.The constructed model had a much shorter detection time than the Faster R-CNN model(only one third of that of the latter),thus meeting the requirements of real-time detection.Conclusion The proposed carotid artery plaque detection method has good performance for accurate identification of carotid plaques in ultrasound images.

Objective The prediction of RR intervals in hypertensive patients can help clinicians to analyze and warn patients'heart condition.Methods Using 8 patients'data as samples,the RR intervals of patients were predicted by long short-term memory network(LSTM)and gradient lift tree(XGBoost),and the prediction results of the two models were combined by the inverse variance method to overcome the disadvantage of single model prediction.Results Compared with the single model,the proposed combined model had a different degree of improvement in the prediction of RR intervals in 8 patients.Conclusion LSTM-XGBoost model provides a method for predicting RR intervals in hypertensive patients,which has potential clinical feasibility.

Objective To develop a method for accurate identification of multiscale carotid plaques in ultrasound images.Methods We proposed a two-stage carotid plaque detection method based on deep convolutional neural network(SM-YOLO).A series of algorithms such as median filtering,histogram equalization,and Gamma transformation were used to preprocess the dataset to improve image quality.In the first stage of the model construction,a candidate plaque set was built based on the YOLOX_l target detection network,using multiscale image training and multiscale image prediction strategies to accommodate carotid artery plaques of different shapes and sizes.In the second stage,the Histogram of Oriented Gradient(HOG)features and Local Binary Pattern(LBP)features were extracted and fused,and a Support Vector Machine(SVM)classifier was used to screen the candidate plaque set to obtain the final detection results.This model was compared quantitatively and visually with several target detection models(YOLOX_l,SSD,EfficientDet,YOLOV5_l,Faster R-CNN).Results SM-YOLO achieved a recall of 89.44%,an accuracy of 90.96%,a F1-Score of 90.19%,and an AP of 92.70%on the test set,outperforming other models in all performance indicators and visual effects.The constructed model had a much shorter detection time than the Faster R-CNN model(only one third of that of the latter),thus meeting the requirements of real-time detection.Conclusion The proposed carotid artery plaque detection method has good performance for accurate identification of carotid plaques in ultrasound images.

Objective To evaluate early outcomes of transthoracic pulmonary valve implantation for the treatment of moderate and severe pulmonary regurgitation by using homemade self-expanding valve (SalusTM). Methods Patients with severe pulmonary regurgitation who underwent transthoracic pulmonary valve implantation in Guangdong Provincial People’s Hospital from September 2, 2021 to November 25, 2022 were prospectively enrolled. The early postoperative complications and improvement of valve and heart function were summarized and analyzed. Results A total of 25 patients were enrolled, including 16 males and 9 females, with an average age of 24.5±1.5 years and an average weight of 57.0±3.0 kg. The mean systolic diameters of the bifurcation near the main pulmonary artery, the stenosis of the middle segment of the aorta and near the valve of the right ventricular outflow tract of the patients were 31.8±7.4 mm, 30.6±5.9 mm and 38.4±8.0 mm, respectively. All patients were successfully implanted with valves, and there were no serious complications such as death, coronary compression, stent fracture, valve displacement and infective endocarditis in the early postoperative period. The indexed left atrial longitudinal diameter, indexed right atrial longitudinal diameter, and indexed right ventricular outflow tract anteroposterior diameter decreased significantly after the operation. The degree of tricuspid and pulmonary valve regurgitation and the indexed regurgitation area decreased significantly. The above differences were statistically significant (P<0.05). Conclusion The early outcomes of transthoracic pulmonary valve implantation with homemade self-expanding pulmonary valve (SalusTM) in the treatment of severe pulmonary regurgitation is relatively good, and the long-term outcomes need to be verified by the long-term follow-up studies with large samples.

Objective To analyze and summarize the early and medium-term outcomes of self-expanding interventional pulmonary valve stent (SalusTM) for right ventricular outflow tract dysfunction with severe pulmonary valve regurgitation. Methods We established strict enrollment and follow-up criteria. Patients who received interventional pulmonary valve in transthoracic implantation in Guangdong Provincial People’s Hospital from September 2, 2021 to July 18, 2023 were prospectively included, and all clinical data of patients were collected and analyzed. Results A total of 38 patients with severe pulmonary regurgitation were included, with 23 (60.5%) males and 15 (39.5%) females. The mean age was 24.08±8.12 years, and the mean weight was 57.66±13.54 kg. The preoperative mean right ventricular end-diastolic volume index (RVEDVI) and right ventricular end-systolic volume index (RVESVI) were 151.83±42.84 mL/m2 and 83.34±33.05 mL/m2, respectively. All patients successfully underwent transcatheter self-expandable pulmonary valve implantation, with 3 (7.9%) patients experiencing valve stent displacement during the procedure. Perioperative complications included 1 (2.6%) patient of postoperative inferior wall myocardial infarction and 1 (2.6%) patient of poor wound healing. The median follow-up time was 12.00 (6.00, 17.50) months. During the follow-up period, there were no deaths or reinterventions, and no patients had recurrent severe pulmonary regurgitation. Three (7.9%) patients experienced chest tightness and chest pain, and 1 (2.6%) patient developed frequent ventricular premature beats. Compared with preoperative values, the right atrial diameter, right ventricular diameter, and tricuspid annular plane systolic excursion were significantly reduced at 6 months and 1 year postoperatively, with improvement in the degree of pulmonary regurgitation (P<0.01). Compared with preoperative values, RVEDVI and RVESVI decreased to 109.51±17.13 mL/m2 and 55.88±15.66 mL/m2, respectively, at 1 year postoperatively (P<0.01). Conclusion 　Self-expanding interventional pulmonary valve in transthoracic implantation is safe and effective for severe pulmonary valve regurgitation and shows good clinical and hemodynamic results in one-year outcome.