Objective:To compare and analyze the parameters of image quality of virtual mono-energetic images(VMIs)and conventional computed tomography(CT)images of dual-energy CT(DECT)imaging for the coracoclavicular ligament,and to optimize the image quality of DECT imaging for coracoclavicular ligament.Methods:The DECT image data of shoulder joints of 50 patients who received DECT scan in the First People's Hospital Affiliated to Shanghai Jiao Tong University were retrospectively analyzed.450 VMIs at energy levels of 40 to 120 keV were obtained through the Syngo.via VB20 workstation,and these VMIs were divided into three groups:low energy level(40,50,60 keV),medium energy level(70,80,90 keV),and high energy level(100,110,120 keV).The contrast-to-noise ratio(CNR),the direction of coracoclavicular ligament's rows,and the displayed subjective quality scores of attachment points of VMIs were compared with those of conventional CT images.The results of MRI examination were used as the"gold standard"to assess the effectiveness of VMIs in diagnosing coracoclavicular ligament injuries of patients with acromioclavicular joint dislocation.The diagnostic efficacies of VMIs'the three groups and conventional images were assessed by using receiver operating characteristic(ROC)curve.Results:The CNR values of diagonal ligament and conical ligament at 40 keV of low energy level group of VMIs for coracoclavicular ligament were respectively 14.09±3.27 and 13.9±3.06,and those of medium energy level group were respectively 11.89±4.01 and 12.17±4.08,and those of high energy level were respectively 9.69±4.25 and 9.78±4.33.The CNR value of low energy level group was significantly higher than that of medium and high energy level groups,and the differences were statistically significant(t=21.1,16.3,P<0.05),respectively.The results of ROC curve analysis indicated that the sensitivity,specificity and area under curve(AUC)value of the low energy level group were relatively higher in diagnosing coracoclavicular ligament injury,which were respectively 90.05%,91.95%and 0.965.There was high consistence in assessing coracoclavicular ligament injury between DECT's VMIs and MRI examination(Kappa=0.896).Conclusion:DECT's VMIs can better display the rows and attachment points of diagonal ligament and conical ligament,and the VMIs quality at low energy level(40 to 70 keV)is the best,which has better diagnostic efficacy in diagnosing coracoclavicular ligament injury.

Objective:To assess potential effect of visual frequency detection technique on the image quality of cardiac magnetic resonance(CMR)cine sequences and its measurement values for myocardial strain.Methods:A retrospective study was conducted on 74 patients without a history of myocardial infarction who underwent CMR scanning at Shanghai General Hospital of Shanghai Jiao Tong University from December 2021 to April 2024.Two 3.0T magnetic resonance imaging(MRI)equipment(equipment A and equipment B)that were different models were used for CMR scanning.Equipment A scanned 38 patients,which used the gain frequency(f)0 Hz of default synthesizer of equipment as center frequency of scan.Equipment B scanned 36 patients,which optimized the frequency of synthesizer by using detection technique for frequency.Subjective evaluation of the image quality of cine sequences for 76 patients was conducted by two physicians.Using image post-processing function of the Cardiovascular Imaging(CVI)42 software to measure the average signal strength of the left ventricular myocardium and blood pool in the cine sequence,and to calculate the signal-to-noise ratio(SNR)and contrast to noise ratio(CNR)of the image,and to compare the differences in myocardial strain values between the measurement of automatic delineation with artificial intelligence(AI)and the measurement of delineation by physicians with AI assistant in the scanned cine sequences of same equipment,and the differences in that of measured values with AI between different equipment,and the differences in that of measured values of physicians with AI between different equipment.Results:In the subjective evaluation of image quality,equipment B appeared non resonant artifact free in 31 patients,which was higher than the number of equipment A(19 cases),and the difference was statistically significant(Z=-3.337,P<0.05).There was no statistically significant difference in the circumferential,radial and longitudinal myocardial strain values that were measured by cine sequence of equipment B between AI and AI combined with physician(P>0.05).The longitudinal myocardial strain value that was measured by AI in the cine sequence of equipment A was 15.27%,and that of AI combined with physician was-13.54%,with a statistically significant difference(x2=2.875,P<0.05).Conclusion:Frequency detection technique can lead to positive influence on the consistency and accuracy of the optimal end diastolic phase,and inner and outer membrane of ventricular wall between the delineation of AI automatic identification and the delineation of the identification of AI combined with physicians through improves the image quality of CMR cine sequences,and increases tissue contrast and contour clarity,which can enhance the accuracy of AI in measuring myocardial strain value.

Objective:To assess potential effect of visual frequency detection technique on the image quality of cardiac magnetic resonance(CMR)cine sequences and its measurement values for myocardial strain.Methods:A retrospective study was conducted on 74 patients without a history of myocardial infarction who underwent CMR scanning at Shanghai General Hospital of Shanghai Jiao Tong University from December 2021 to April 2024.Two 3.0T magnetic resonance imaging(MRI)equipment(equipment A and equipment B)that were different models were used for CMR scanning.Equipment A scanned 38 patients,which used the gain frequency(f)0 Hz of default synthesizer of equipment as center frequency of scan.Equipment B scanned 36 patients,which optimized the frequency of synthesizer by using detection technique for frequency.Subjective evaluation of the image quality of cine sequences for 76 patients was conducted by two physicians.Using image post-processing function of the Cardiovascular Imaging(CVI)42 software to measure the average signal strength of the left ventricular myocardium and blood pool in the cine sequence,and to calculate the signal-to-noise ratio(SNR)and contrast to noise ratio(CNR)of the image,and to compare the differences in myocardial strain values between the measurement of automatic delineation with artificial intelligence(AI)and the measurement of delineation by physicians with AI assistant in the scanned cine sequences of same equipment,and the differences in that of measured values with AI between different equipment,and the differences in that of measured values of physicians with AI between different equipment.Results:In the subjective evaluation of image quality,equipment B appeared non resonant artifact free in 31 patients,which was higher than the number of equipment A(19 cases),and the difference was statistically significant(Z=-3.337,P<0.05).There was no statistically significant difference in the circumferential,radial and longitudinal myocardial strain values that were measured by cine sequence of equipment B between AI and AI combined with physician(P>0.05).The longitudinal myocardial strain value that was measured by AI in the cine sequence of equipment A was 15.27%,and that of AI combined with physician was-13.54%,with a statistically significant difference(x2=2.875,P<0.05).Conclusion:Frequency detection technique can lead to positive influence on the consistency and accuracy of the optimal end diastolic phase,and inner and outer membrane of ventricular wall between the delineation of AI automatic identification and the delineation of the identification of AI combined with physicians through improves the image quality of CMR cine sequences,and increases tissue contrast and contour clarity,which can enhance the accuracy of AI in measuring myocardial strain value.

Objective:To compare and analyze the parameters of image quality of virtual mono-energetic images(VMIs)and conventional computed tomography(CT)images of dual-energy CT(DECT)imaging for the coracoclavicular ligament,and to optimize the image quality of DECT imaging for coracoclavicular ligament.Methods:The DECT image data of shoulder joints of 50 patients who received DECT scan in the First People's Hospital Affiliated to Shanghai Jiao Tong University were retrospectively analyzed.450 VMIs at energy levels of 40 to 120 keV were obtained through the Syngo.via VB20 workstation,and these VMIs were divided into three groups:low energy level(40,50,60 keV),medium energy level(70,80,90 keV),and high energy level(100,110,120 keV).The contrast-to-noise ratio(CNR),the direction of coracoclavicular ligament's rows,and the displayed subjective quality scores of attachment points of VMIs were compared with those of conventional CT images.The results of MRI examination were used as the"gold standard"to assess the effectiveness of VMIs in diagnosing coracoclavicular ligament injuries of patients with acromioclavicular joint dislocation.The diagnostic efficacies of VMIs'the three groups and conventional images were assessed by using receiver operating characteristic(ROC)curve.Results:The CNR values of diagonal ligament and conical ligament at 40 keV of low energy level group of VMIs for coracoclavicular ligament were respectively 14.09±3.27 and 13.9±3.06,and those of medium energy level group were respectively 11.89±4.01 and 12.17±4.08,and those of high energy level were respectively 9.69±4.25 and 9.78±4.33.The CNR value of low energy level group was significantly higher than that of medium and high energy level groups,and the differences were statistically significant(t=21.1,16.3,P<0.05),respectively.The results of ROC curve analysis indicated that the sensitivity,specificity and area under curve(AUC)value of the low energy level group were relatively higher in diagnosing coracoclavicular ligament injury,which were respectively 90.05%,91.95%and 0.965.There was high consistence in assessing coracoclavicular ligament injury between DECT's VMIs and MRI examination(Kappa=0.896).Conclusion:DECT's VMIs can better display the rows and attachment points of diagonal ligament and conical ligament,and the VMIs quality at low energy level(40 to 70 keV)is the best,which has better diagnostic efficacy in diagnosing coracoclavicular ligament injury.

Objective:To compare the biomechanical properties of posterolateral distal humeral plate, inverted anterior proximal humerus internal locking system (PHILOS), and anterior reconstruction plate in the treatment of distal humeral shaft fractures by a finite element analysis.Methods:One healthy adult male volunteer, aged 27 years, with a height of 171 cm and a weight of 70 kg, was recruited for this study. The finite element method was used to establish a simulation model of distal humeral shaft fracture. The maximum displacement and maximum stress were compared between fixation with posterolateral distal humeral plate (group A), fixation with inverted anterior PHILOS (group B), and fixation with anterior reconstruction plate (group C).Results:In groups A, B, and C, respectively, the overall stress peak values were 409.07 MPa, 217.04 MPa, and 370.64 MPa; the peak stresses under torsional load were 234.55 MPa, 348.80 MPa and 458.17 MPa; the overall stress peaks under bending load were 250.22 MPa, 466.76 MPa, and 582.32 MPa. The smaller the stress, the smaller the risk of fatigue fracture. In groups A, B, and C, respectively, the overall displacement peak values were 5.18 mm, 3.04 mm and 3.75 mm; the peak displacements under torsional load were 1.20 mm, 1.02 mm and 2.05 mm; the peak displacements under bending load were 3.85 mm, 5.28 mm and 9.04 mm. The smaller the displacement, the better the fixation stability.Conclusions:In the treatment of distal humeral shaft fractures, fixation with inverted anterior PHILOS leads to the best mechanical stability under axial compression and torsional stress, while fixation with the posterolateral distal humeral plate leads to the best mechanical stability under bending stress.