Objective To develop a localization strategy for magnetic resonance coronary angiography (MRCA). Methods In 89 subjects, the standard 4-chamber view and long-axis view of left and right ventricle were acquired using Fast-Imaging-Employing-Steady-State-Acquisition (FIESTA) sequence in CINE mode, and the trigger-delay time for mid-diastolic phase was determined. Coronary vessels including right coronary artery (RCA), left main (LM), left anterior descending (LAD), and left circumflex (LCX) were localized and imaged using 3-dimensional fat-suppressed FIESTA sequence during end-expiration. The reproducibility of the localization strategy was evaluated by taking the standard of coronary segmentation system recommended by American Heart Association. Results Eighty-six subjects completed the examination with full respiratory co-operation and the indication ratio was 96.63%. Nine planes were optimized as the standard to target the main branches of coronary arteries, and a comprehensive reproducibility reached 100% in demonstrating the proximal and middle segment of RCA (AHA-18, 19), LM (AHA-1, 2), proximal and middle segment of LAD (AHA-3, 5, 7), and proximal LCX (AHA-10). The reproducibility for the demonstration of distal segments of LAD, LCX, and RCA (AHA-9, 14, 21) was 94.19%, 72.09%, and 96.51%, respectively. Conclusion This is a simple and practical localization strategy for MRCA. It could image the proximal and middle segments of the coronary arteries with good reproducibility, which indicates the potential for clinical application.

Objective To investigate the spiral CT findings in hemopathic patients with druginduced pulmonary injury.Methods CT images obtained in 11patients with drug-induced pulmonary injury were retrospectively analyzed.Six patients had antineoplastic agent-induced pulmonary injury and 5 patients had non-neoplastic agent-induced pulmonary injury (immunosuppressor in 2 patients,antifungal in 2 patients,antineoplastic immunomodulators in 1 patient).CT findings were reviewed by a chest radiologist.Results All 11patients had parenchymal abnormalities on MSCT scans,including ground-glass opacities( n =8 ),consolidation( n =5 ),interlobular septal thickening( n =3 ) and focal fibrosis ( n =2 ).The abnormalities were bilateral and asymmetric in all patients.They were mainly in the peripheral lung regions in 6 patients,in the central lung regions in four,and irregularly located in one.The abnormalities involved mainly the lower lung zones in six patients,the upper lung zones in two,and all lung zones homogeneously in three.One patient had fluid in bilateral pleural cavities.Three patients were given the same agent once more after the imaging turned to normal,and they presented with same clinical symptoms and similar but more serious imaging findings.Conclusions Drug-induced pulmonary injury usually manifests as areas of ground-glass opacity and consolidation,which most commonly involves the peripheral lungs and lower lung zones.Drug-induced pulmonary injury shows reproducible but more serious lesions when the patient is given the same agent once more.

Objective To investigate the changes of quantitative parameters of dynamic enhanced CT in non-small cell lung cancer before and after targeted therapy,and compare them with the traditional evaluation criteria,in order to find the parameters which can be exploited for timely,objective evaluation of the effect of targeted therapy.Methods The study included 21 patients with targeted therapy who had received dynamic enhanced CT before and after treatment.Enhancement time-density curves were obtained based on the CT values of the lesion at individual time points,and the functional indices:peak height (PH),the time to peak height (Tp),the ratio of PH of the mass to aorta (M/A) and perfusion value were calculated.The effects of the treatment on these indices were evaluated and compared with the effect of the treatment on lesion diameter. Results Twenty-one patients had 33 rechecking results. There was a statistically significant agreement between lesion diameter-based treatment evaluation and perfusion-based treatment evaluation ( U =8.761,P ＜ 0.01 ). The perfusion value decreased in patients with disease regression[before treatment:(0.28 ±0.11 ) ml · min-1 · ml-1,after targeted therapy(0.18 ±0.09) ml ·min-1 · ml-1,t =- 3.2722,P =0.0042],but increased in patients with disease progression[before treatment(0.21 ±0.08) ml · min-1 · ml-1,after targeted therapy:(0.34 ±0.11 ) ml · min-1 · ml-1,t =2.6064,P =0.0403].Conclusions On dynamic enhanced CT in non-small cell lung cancer patients after targeted therapy,perfusion value changed in the same trend as the diameter of tumor.The effectiveness of targeted therapy may be evaluated by perfusion value changes.

Objective To analyze the characteristics of breast cancer blood supply before and after chemotherapy with low-dose CT perfnsion. Methods Fifteen patients with breast cancer underwent CT breast perfusion examination, which was performed before and after chemotherapy within 1 week on Siemens Sensation 4 scanner with 120 kV and 50 mAs, 50 ml of nonionic contrast agent (320 mg I/ml) was injected at a flow rate of 4 ml/s with a power injector, Scan started after 8 seconds delay and data acquisition duration was 50 seconds. The blood flow ( BF), blood volume (BV) and mean transfer time (MTT) of lesion and contralateral normal breast gland were calculated using Basama perfusion 3 software package before and after chemotherapy. At the same time, the tumor size before and after chemotherapy were measured and correlated with the BF values. The t test and non-parametric test were used for the statistics. Results ( 1 ) The mean BF、BV and MTT of breast cancer were (33.20±4. 17) ml · min-1 · 100 ml-1 , (8. 31±2.43) ml · 100 ml-1 and ( 15. 31 ± 4. 31 ) s respectively before chemotherapy, and ( 13.65 ± 6. 04) ml · min-1 100 ml-1, (5.04±2. 33) ml · 100 ml-1 and (25. 97±9. 07) s respectively after chemotherapy and there were statistically significant (P =0. 000). The mean BF、BV and MTT of normal breast were (4. 31 ± 2.23) ml · min-1 · 100 ml-1, (1.38±0.75) ml · 100 ml-1 and (19.25±3.94) s respectively before chemotherapy, and (4.03±2.35) ml · min-1 · 100 ml-1、(1.44±0.84) ml · 100 ml-1、(22.56 ± 7.71 ) s respectively after chemotherapy and there were not statistically significant (P >0. 05). (2)The BF of breast cancer was higher than the normal breast before chemotherapy ( P < 0. 01 ). (3) There was a positive correlation between the BF values and tumor size before and after chemotherapy ( r = 0. 902, P = 0. 000). Conclusion The BF value has a positive correlation with tumor size after chemotherapy, CT perfusion is more sensitive for the evaluation of chemotherapy response than morphologic assessment.

Objective To evaluate the efficiency of coronary magnetic resonance angiography (CMRA) for stenoses detection by using breath-hold three-dimensional fast imaging employing steady state acquisition (FIESTA) sequence with the reference of conventional coronary catheter angiography. Methods~Consecutive 33 patients accepted CMRA examination within 3 weeks after the catheter angiography. Coronary stenoses was graded in 5 levels as 0%, 0%-25%, 25%-50%, 50%-75%, and 75%-100%, respectively, and CMRA and catheter angiogram were compared segment by segment. Results For the differentiation of the stenoses 50%, the accuracy, sensitivity, and specificity of CMRA was 84.3%, 84.8%, and 84.1%, respectively, and the negative prediction value was 92.3%. The accuracy, sensitivity, and specificity for the differentiation of stenoses between 50%-75% and 75%-100% were all 61.5%. Conclusion The breath-holding three-dimensional FIESTA sequence for CMRA was practical to exclude hemodynamic significant coronary stenoses but limited in detail grading.

Objective To analyze the imaging characteristics of thoracic LDRD(Low-dose directly Digital Radiographic Device) and its artifacts.Methods 188 patients were performed with LDRD and common thoracic X-ray film respectively in our hospital during two weeks.Results Among the 188 cases,1.60%(3/188) showed thoracic motion artifacts.46.8%(88/188) appeared as tentorial prominence along left heart edge and 2.6%(5/188) along the right one.1 artifact was in aorta-pulmonary artery window(0.53%).Conclusion(1)Less than 0.5 should be taken as reference in podoid enlargement diagnosis.(2)Pseudomorph from heart motion may result from cardio-phase,cardiac contraction,heart rate,arrhythmia,local abnormal pulse of left heart edge,different enlarged velocity of cardiac cavity during heart beat,etc.(3)The motion artifacts in thoracic LDRD has no important influence in clinical diagnosis and therapy.

Objective: To evaluate the performance of a deep learning (DL) based mammogram calcification detection system. Methods: Screening digital mammographic examinations with standard cranio-caudal (CC) and medio-lateral oblique (MLO) views were performed in 1 431 women (5 488 mammogram images) who were enrolled between January and December in 2013. The DL system and a radiologist detect calcifications separately, and then both results are reviewed by a moreexperiencedradiologist. Sensitivities of the DL model and radiologist were compared. Different calcification morphology, distribution, BI-RADS categories, breast density and patient age were investigated by χ² tests. Results: For DL system, sensitivity of all kinds of calcifications were 96.76% (7 649/7 905). The average false positive was 1.04 per image (5 706/5 488), 3.99 per case (5 706/1 431). The false positive rate was 42.73% (5 706/13 355). There was no significant differences for DL system with different calcification distribution, BI-RADS categories, breast densities and patient ages (P>0.05). Conclusion Deep learning based mammogram calcification detection system shows high sensitivity and stability, which may help to reduce the missing rate of calcification (especially the suspicious ones).