Objective To evaluate the application value of artificial intelligence (AI) in the quality control of mammographic images and explore its feasibility for improving image quality. Methods A retrospective analysis was conducted on 500 mammographic images from 125 female patients. These images were acquired in December 2024 at the Department of Radiology, Obstetrics and Gynecology Hospital of Tongji University by two junior technologists with qualification certificate for junior radiologic technologists and ≤ 2 months of independent operation. The reference standard was the evaluation by a panel of senior experts comprising one associate chief radiologist and two intermediate radiologic technologists with over 10 years of experience in mammography. The evaluation was based on nine criteria within an AI quality control system. The correlation and consistency (Kappa test) of the assessment outcomes were compared among the junior technologist group, the AI quality control group, and the panel group. Additionally, the differences in the proportions of high-, medium-, and low-quality images rated by the three groups were analyzed. After one month of AI assistance, 200 images from 50 additional patients acquired by the same two junior technologists in January 2025 were collected. The area under the receiver operating characteristic curve and 95% confidence interval (95%CI) were calculated for the assessment performance of the junior technologist group before and after AI assistance. Differences were compared using the DeLong test. Changes in the disqualification rates for the nine criteria were analyzed. Results In consistency analysis, the AI quality control group showed high consistency with the panel group across multiple key criteria, with Kappa values ranging from 0.41 to 1.00. In contrast, the consistency between the junior technologist group and the panel group was generally low, with Kappa values ranging from 0.13 to 0.49. In comparison of image quality classification, no significant differences were observed in image quality classification between the AI group and the panel group (P>0.05). However, the proportion of images rated as high quality by the junior technologist group was significantly higher, while the proportions rated as medium and low quality were significantly lower, compared to those rated by the panel group (P<0.05). After AI assistance, the area under the receiver operating characteristic curve for the assessment performance of junior technologist group increased significantly from 0.56 (95%CI: 0.51-0.61) to 0.91 (95%CI: 0.87-0.94) (P<0.001). The disqualification rates for skin folds, incomplete inclusion of the pectoralis major muscle, and nipple not in profile decreased significantly. Conclusion AI demonstrates assessment capability comparable to the panel group in mammographic quality control. AI effectively enhances the quality of images acquired by junior technologists through real-time, objective feedback.

Objective To evaluate the application value of artificial intelligence (AI) in the quality control of mammographic images and explore its feasibility for improving image quality. Methods A retrospective analysis was conducted on 500 mammographic images from 125 female patients. These images were acquired in December 2024 at the Department of Radiology, Obstetrics and Gynecology Hospital of Tongji University by two junior technologists with qualification certificate for junior radiologic technologists and ≤ 2 months of independent operation. The reference standard was the evaluation by a panel of senior experts comprising one associate chief radiologist and two intermediate radiologic technologists with over 10 years of experience in mammography. The evaluation was based on nine criteria within an AI quality control system. The correlation and consistency (Kappa test) of the assessment outcomes were compared among the junior technologist group, the AI quality control group, and the panel group. Additionally, the differences in the proportions of high-, medium-, and low-quality images rated by the three groups were analyzed. After one month of AI assistance, 200 images from 50 additional patients acquired by the same two junior technologists in January 2025 were collected. The area under the receiver operating characteristic curve and 95% confidence interval (95%CI) were calculated for the assessment performance of the junior technologist group before and after AI assistance. Differences were compared using the DeLong test. Changes in the disqualification rates for the nine criteria were analyzed. Results In consistency analysis, the AI quality control group showed high consistency with the panel group across multiple key criteria, with Kappa values ranging from 0.41 to 1.00. In contrast, the consistency between the junior technologist group and the panel group was generally low, with Kappa values ranging from 0.13 to 0.49. In comparison of image quality classification, no significant differences were observed in image quality classification between the AI group and the panel group (P>0.05). However, the proportion of images rated as high quality by the junior technologist group was significantly higher, while the proportions rated as medium and low quality were significantly lower, compared to those rated by the panel group (P<0.05). After AI assistance, the area under the receiver operating characteristic curve for the assessment performance of junior technologist group increased significantly from 0.56 (95%CI: 0.51-0.61) to 0.91 (95%CI: 0.87-0.94) (P<0.001). The disqualification rates for skin folds, incomplete inclusion of the pectoralis major muscle, and nipple not in profile decreased significantly. Conclusion AI demonstrates assessment capability comparable to the panel group in mammographic quality control. AI effectively enhances the quality of images acquired by junior technologists through real-time, objective feedback.

Objective To evaluate the application value of artificial intelligence (AI) in the quality control of mammographic images and explore its feasibility for improving image quality. Methods A retrospective analysis was conducted on 500 mammographic images from 125 female patients. These images were acquired in December 2024 at the Department of Radiology, Obstetrics and Gynecology Hospital of Tongji University by two junior technologists with qualification certificate for junior radiologic technologists and ≤ 2 months of independent operation. The reference standard was the evaluation by a panel of senior experts comprising one associate chief radiologist and two intermediate radiologic technologists with over 10 years of experience in mammography. The evaluation was based on nine criteria within an AI quality control system. The correlation and consistency (Kappa test) of the assessment outcomes were compared among the junior technologist group, the AI quality control group, and the panel group. Additionally, the differences in the proportions of high-, medium-, and low-quality images rated by the three groups were analyzed. After one month of AI assistance, 200 images from 50 additional patients acquired by the same two junior technologists in January 2025 were collected. The area under the receiver operating characteristic curve and 95% confidence interval (95%CI) were calculated for the assessment performance of the junior technologist group before and after AI assistance. Differences were compared using the DeLong test. Changes in the disqualification rates for the nine criteria were analyzed. Results In consistency analysis, the AI quality control group showed high consistency with the panel group across multiple key criteria, with Kappa values ranging from 0.41 to 1.00. In contrast, the consistency between the junior technologist group and the panel group was generally low, with Kappa values ranging from 0.13 to 0.49. In comparison of image quality classification, no significant differences were observed in image quality classification between the AI group and the panel group (P>0.05). However, the proportion of images rated as high quality by the junior technologist group was significantly higher, while the proportions rated as medium and low quality were significantly lower, compared to those rated by the panel group (P<0.05). After AI assistance, the area under the receiver operating characteristic curve for the assessment performance of junior technologist group increased significantly from 0.56 (95%CI: 0.51-0.61) to 0.91 (95%CI: 0.87-0.94) (P<0.001). The disqualification rates for skin folds, incomplete inclusion of the pectoralis major muscle, and nipple not in profile decreased significantly. Conclusion AI demonstrates assessment capability comparable to the panel group in mammographic quality control. AI effectively enhances the quality of images acquired by junior technologists through real-time, objective feedback.

OBJECTIVE: To exploit material decomposition analysis in dual-energy spectral computed tomography (CT) to assess the blood supply status of the ground-glass opacity (GGO) in lungs. MATERIALS AND METHODS: This retrospective study included 48 patients with lung adenocarcinoma, who underwent a contrast-enhanced dual-energy spectral CT scan before treatment (53 GGOs in total). The iodine concentration (IC) and water content (WC) of the GGO, the contralateral and ipsilateral normal lung tissues were measured in the arterial phase (AP) and their differences were analyzed. IC, normalized IC (NIC), and WC values were compared between the pure ground-glass opacity (pGGO) and the mixed ground-glass opacity (mGGO), and between the group of preinvasive lesions and the minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IA) groups. RESULTS: The values of pGGO (IC = 20.9 ± 6.2 mg/mL and WC = 345.1 ± 87.1 mg/mL) and mGGO (IC = 23.8 ± 8.3 mg/mL and WC = 606.8 ± 124.5 mg/mL) in the AP were significantly higher than those of the contralateral normal lung tissues (IC = 15.0 ± 4.9 mg/mL and WC = 156.4 ± 36.8 mg/mL; IC = 16.2 ± 5.7 mg/mL and WC = 169.4 ± 41.0 mg/mL) and ipsilateral normal lung tissues (IC = 15.1 ± 6.2 mg/mL and WC = 156.3 ± 38.8 mg/mL; IC = 15.9 ± 6.0 mg/mL and WC = 174.7 ± 39.2 mg/mL; all p < 0.001). After normalizing the data according to the values of the artery, pGGO (NIC = 0.1 and WC = 345.1 ± 87.1 mg/mL) and mGGO (NIC = 0.2 and WC = 606.8 ± 124.5 mg/mL) were statistically different (p = 0.049 and p < 0.001, respectively), but not for the IC value (p = 0.161). The WC values of the group with preinvasive lesions and MIA (345.4 ± 96.1 mg/mL) and IA (550.1 ± 158.2 mg/mL) were statistically different (p < 0.001). CONCLUSION: Using dual-energy spectral CT and material decomposition analysis, the IC in GGO can be quantitatively measured which can be an indicator of the blood supply status in the GGO.
Adenocarcinoma ; Arteries ; Humans ; Iodine ; Lung ; Multidetector Computed Tomography ; Retrospective Studies ; Tomography, X-Ray Computed ; Water

Purpose To investigate the clinicopathological features of multilocular cystic renal cell carcinoma ( MCRCC) , and to im-prove the understanding of this disease. Methods Thirty-two of MCRCC were studied by clinic data, pathological features and immu-nophenotype. All the thirty-two cases were followed up. Results In this study, thirty-two patients were diagnosed as MCRCC with a male-to-female ratio of 2.2 : 1, thirteen of whom had the tumor in the left kidney, eighteen in the right kidney, another in the double kidneys. The mean of tumor diameter was 4.6 cm (1.0~8.0 cm). Eleven patients underwent radical nephrectomy while the other 21 patients received nephron sparing surgery. Microscopically, all the cases were multiocular, lined in the cyst wall by a single layer of tumor cells with the clear or pale cytoplasm and Fuhrman grade 1 nuclei. Occasionally, the lining consisted of several layers of tumor cells or a few small papillae were present. The linings of the cyst wall were rich in thin-wall blood vessels. Immunohistochemically, the tumor cells were positive for CK(32/32), CK7(25/32), EMA(32/32), CD10(23/32) and vimentin(20/32), while negative for CD68. Conclusions MCRCC is characteristic by low grade nuclei, lacking solid nodules, and in the cyst wall, and has a favourable prognosis. The rich in thin-wall blood vessels and the lining cells with the clear or pale cytoplasm are diagnosis clues.

Objective To assess the clinical value of volume helical shuttle (VHS) of highdefinition CT (HDCT) in the diagnosis of congenital heart disease (CHD).Methods Forty-three patients suffering from CHD were examined with VHS of HDCT and echocardiography (ECHO).Twenty-four patients (21 children,3 adults) received surgery.Based on surgical data,the diagnostic accuracy of CHD by VHS of HDCT was compared with that by ECHO with respect to the intra-cardiac lesion,extra-cardiac lesion and hemodynamics.The visibility of the origin and proximal course of coronary arteries was also evaluated on VHS.CT effective doses were calculated,using mSv =DLP × K. Results Thirty-six intra-cardiac deformities were found by operation,and the diagnostic accuracy of both VSH and ECHO were 100.0％ (36/36).VHS successfully detected 42 of 44 (95.4％) cardiac-vascular connection and large vascular deformities,whereas ECHO only detected 26 (59.1％ ).The right-to-left shunt and left-to-right shunt of septal defects revealed by VHS were 5 and 6 patients respectively,which all coincided with the results of ECHO.Fourteen bi-directional shunt of septal defects were detected by ECHO,and 13 of them were diagnosed as right-to-left shunt by HDCT except an illegibility in 1 patient.The origin and proximal course of coronary arteries were all visible on VHS.Effective dose for children was (3.38 ± 1.86) mSv,and for adults was ( 10.27 ± 2.59) mSv.Conclusions VHS of HDCT scan is advantageous in the diagnosis of CHD.It can depict the intra- and extra- cardiac structures as well as coronary arteries and provide hemodynamic information.

Objective To analyze the distribution of image noise in low-dose chest CT scan and optimize the relative scanning parameters.Methods The CT images of the Chinese anthropomorphic chest phantom( CDP-1 C) were simulated into six groups of low-dose images with different noise indexs by using an image noise addition tool.The difference between the preset noise index and analog noise value was compared.The CT images of 20 volunteers were also simulated into nine groups of low dose scans with the tube currents of 10,30,50,80,100,120,150,180 and 240 mA.The noise values of images were recorded and analyzed.Results There was no statistical difference between the analog noise value and the noise index.The image noise of low-dose chest scan was increased with the decrease of tube current.The noise was increased quickly when the current was decreased from 50 to 30 mA ( F =24.09 - 40.79,P ＜ 0.05),but the noise increased slowly when the current decreased from 240 to 80 mA.There was no statistical difference between the noise of 80 mA group and that of 120 mA(P ＞ 0.05).Conclusions The noise addition tool can be used to evaluate the image noise of low-dose chest CT scan.Adoption of 80 mA in chest CT scan would result in low radiation dose without adding image noise.

Objective To assess the capability of monochromatic energy images of Gemstone spectral imaging(GSI) in reducing artifacts caused by metallic implants. Methods Twelve subjects with metallic implants underwent GSI (Discovery CT750 HD, GE Healthcare, Milwaukee ). The metallic orthopedic implants included 3 patients of dentures, 2 patients of cervical spinal vertebraplasty, one clavicle fracture fixation, one lumbar spinal vertehraplasty, 3 patients of artificial femoral head, one iliac fracture fixation and one tibial fracture fixation. GSI was performed by using a single source ultra-fast dual energy X-ray switching (80 kVp and 140 kVp). Following GSI scanning, thin slice images were reconstructed into 1.25 mm slice thickness. The monochromatic energy images were set to the same window width and level (window width 1500 HU,window level 500 HU). The artifact indexes (AI) at different kiloelectronvolts (keV) images were measured and compared. 3D reconstruction was performed using images with minimal AI. Result The artifacts index on monochromatic energy images varied with the change of keV. Of the images from 12 subjects, the maximal AI ranged between 145-225 at 40 keV, and minimal AI ranged between 15-90 at the 95-140 keV. The artifacts are clearly visible on polychromatic energy images and the artifacts are reduced markedly on the monochromatic energy images with minimal AI. Conclusion The artifacts caused by metallic implants can be reduced significantly by GSI with high keV monochromatic energy images.

Objective To assess the value of spectral CT in reducing artifacts caused by metallic implants. Methods Porcine lumbar spines were chosen as anthropomorphic phantom. The model was examined before and after implanting the titanic nail into the second and fourth lumbar vertebral body using gemstone spectral CT protocol and standard 120 kVp spectra. Specific pest-processing technique was applied to generate 11 kinds of images of monochromatic energy and Metal Artifacts Reducing system (MARs) with the interval of 10 keV ranging from 40-140 keV. The image quality was compared subjectively between 120 kVp group and GSI group after implantation. Three regions of interest based on distances along the most pronounced artifact were chosen and marked as ROInear, ROImid, ROIfar successively. Artifacts parameters including CT value and SD value were measured. The CT value of different ROIs were compared with LSD and Bonferroni test. Contrast-to-noise ratio and artifacts index were calculated. An optimal range of keV was determined according to artifacts index. Results Image quality of Gemstone spectra images was rated superior to the standard images. An optimized spectrum of keV based on artifacts index was from 80 keV to 100 keV. For ROInear,CT value was ( 80. 25 ± 16. 00) HU and ( 30. 10 ± 10. 45 ) HU respectively in group Mono before implantation and group Mono + MARs after implantation. The differences were significant( Z =2. 978, P ＜ 0. 05 ). For ROImid and ROIfax, CT value was ( 63.21 ± 6. 61 ) HU and ( 54. 84 ± 10. 60 ) HU,(76. 54 ±9.07)HU and (73.20 ±5.39)HU respectively. There was no significant differences (t =0. 530,P ＞ 0. 05; t = 0. 822, P ＞ 0. 05 ). Conclusions Metal artifacts could be reduced effectively at the site 3 cm away from implants using gemstone spectral CT. An accurate CT value of surrounding tissue can be obtained.

Objective To investigate the feasibility of differentiating lymph node metastases of four types of primary tumors (lymphoma, lung adenocarcinoma, lung squamous cell carcinoma and cholangiocarcinoma) using gemstone spectral imaging (GSI) . Methods Three cases with lymphoma (28 lymph node), five cases with lung adenocarcinoma(30 lymph node), four cases with lung squamous cell carcinoma(24 lymph node) and two cases with cholangiocarcinoma( 10 lymph node) were evaluated by germstona spectra imaging CT scans. Imaging protocol included unenhanced conventional CT scan (120 kVp) ,enhanced GSI (80/140 kVp) on arterial phase and conventional CT scan (120 kVp) on portal phase. CT attenuation values of lymph nodes in the monochromatic images at 11 sets of keV levels (40-140 keV, 10 keV step) and the iodine and water contents of these lymph nodes were measured. All results were analyzed with ANOVA and t test. Results The optimal monochromatic level was 70 keV for the optimal contrast-noise ratio (CNR) of metastatic lymphadenopathy. The CT attenuation values of metastatic lymphadenopathy were (81.36 ±9. 81 ), (58.33 ± 21.55 ), (56. 47 ± 10.62) and (73. 57 ±4. 43 ) HU,respectively, at 70 keV( F = 17.29, P ＜0. 01 ). There were significant differences in CT attenuation values between lymphoma and lung adenocarcinoma, between lymphoma and lung squamous cell carcinoma and between lung squamous cell carcinoma and cholangiocarcinoma (P ＜ 0. 05 ). The differences in CT attenuation values were significant between cholangiocarcinoma and lung squamous cell carcinoma, between cholangiocarcinoma and lymphoma ( P ＜ 0. 05 ). There was no difference in CT attenuation values at all 11 sets of keV levels between lung squamous cell carcinoma and lung adenocarcinoma ( P ＞ 0. 05 ). The iodine contents of lymphoma, lung adenocarcinoma, lung squamous cell carcinoma and cholangiocarcinoma were ( 1. 93 ± 0. 04 ), ( 1.16 ± 0. 15 ), ( 1.25 ± 0. 21 ) and ( 1.44 ± 0. 04 ) g/L, respectively. The water contents of lymphoma, lung adenocarcinoma, lung squamous cell carcinoma and cholangiocarcinoma were (1029.40 ± 20. 85), (1024.98 ± 11.19), (1022.12 ± 12. 94) and (1030.87 ± 10.10) g/L,respectively. Except between lung squamous cell carcinoma and lung adenocarcinoma, the differences in the iodine contents of metastatic lymphadenopathy were significant among tumors ( P ＜ 0. 05 ). There was no difference in the water contents of metastatic lymphadenopathy among tumors ( P ＞ 0. 05 ). Conclusions Although CT spectral imaging fails to differentiate metastatic lymphadenopathy of lung adenocarcinoma and lung squamous cell carcinoma, it is also a promising method of distinguishing metastatic lymphadenopathy of malignant tumors by CT attenuation values in monochromatic images and iodine contents in material density images. The optimal monochromatic level was determined to be at 70 keV for providing the optimal CNR of metastatic lymphadenopathy.