239 Objectives: Because of the low sensitivity, clinical PET/CT system requires more than 10 minutes to finish a whole body PET scan. Such long time scanning may bring discomfort to the patient and introduce motion artifacts into the reconstructed image. It is urgently needed to reduce the scanning time while the image is still clinically acceptable. In this work, we proposed a deep learning based method for short-time PET imaging and compared it with conventional PET image de-noising methods. The aim of this work is to investigate the possibility to reduce the PET acquisition time by half, while the image quality is still comparable to that from standard acquisition time.
Methods: Two datasets were generated for comparison. The image set A was reconstructed by using the standard protocols adopted in routine clinical practice, and the image set B was reconstructed by using half of the PET acquisition time to simulate the short-time PET imaging. A modified U-Net deep neuro-network was constructed and trained, with image dataset B as input and image dataset A as target. The network structure is displayed in figure 1. Totally 108 clinical images were used for training and other 22 images were used for testing. Also the image set B was processed with conventional image de-noising methods, i.e. Gaussian filtering and non-local mean (NLM) filtering. With image set A as ‘ground truth’, image quality assessment was performed on the images generated from image set B to see the merits of different de-noising methods. Quantitative evaluation included PSNR, NMSE and liver SNR. Additionally, a patch-to-patch comparison method was developed to address the SUV accuracy of small lesions. Images were decomposed into overlapped patches with patch size of 12mm x 12mm x 12mm, and then the mean value of each patch was calculated and compared to the ground truth. Qualitative evaluation was conducted by radiologists with at least 5 years’ experience. The radiologists reviewed the image set A and B in a blind review and were asked to rate a score 1-5 (from the worst to the best) for the lesion detectability. Results: Results of PSNR, NMSE, liver SNR and SUV accuracy are shown in figure 2. Images without any noise reduction techniques gets the lowest score (3) and the Deep learning based method gets the highest score (4.45). The scores of Gaussian filtering, NLM filtering and Ground truth are 4.09, 4.18 and 4.23, respectively.
Conclusions: Gaussian filtering shows the worst de-noising performance on short-time PET imaging. The SUV accuracy of small regions drops a lot because the image is over-smoothed. NLM filtering is better than Gaussian filtering on noise reduction. It preserves the SUV accuracy of small regions before and after noise reduction. Deep learning based method outperforms Gaussian filtering and NLM filtering on both noise reduction and quantitatively accuracy. The increased accuracy of SUV may attribute to the prior information learned in the neuro-network. Deep learning based method also receives highest score from radiologists’ assessment.
Objective The objective of this study was to compare the image quality (IQ), contrast medium (CM) volume, and radiation dose of the high-pitch renal computed tomography angiography (CTA) with low-pitch protocol. Methods Fifty patients underwent renal CTA on a dual-source 128-slice scanner via a high-pitch mode (pitch = 2.05) with 0.5-mL/kg CM injection, whereas 50 patients were also scanned on the same scanner with low-pitch (pitch = 0.6) and 1.0 mL/kg CM injection. Subjective IQ was evaluated. Objective IQ was determined by the signal-to-noise ratio and contrast-to-noise ratio. Effective radiation dose was also evaluated. Results The contrast-to-noise ratio and signal-to-noise ratio values as well as the IQ scores between the 2 groups had no significant differences (P > 0.05). The effective radiation dose of the high-pitch group was significantly lower (P < 0.05). Conclusions High-pitch scan can provide similar subjective and objective IQ compared with low-pitch protocol for renal CTA, whereas CM volume and radiation exposure were significantly reduced.
The aim of this study was to validate quantitative performance of a newly released simultaneous positron emission tomography (PET)/magnetic resonance imaging (MRI) scanner, by using MR-based attenuation correction (MRAC), both in phantom study and in patient study. PET/MRI image uniformities of a phantom under different hardware configurations were tested and compared. Thirty patients were examined with 2-deoxy-2-[ 18 F]fluoro-D-glucose ( 18 F-FDG) PET/computed tomography (CT) and subsequent PET/MRI. PET images from PET/MRI were corrected with MRAC (PET MR ), CT-based attenuation maps ( μ -maps, PET CT ), and segmented CT μ -maps (PET CTSeg ) derived from PET/CT. Standardized uptake values (SUVs) were compared among the 3 sets of PET in main organs (bone, liver and lung) and in 52 FDG-avid lesions, including soft-tissue lesions and bone lesions. The result showed that PET imaging uniformities of PET/MRI under different configurations were good (<8.8%). The SUV differences among the 3 sets of PET varied with organs and lesion types. In detail, the mean relative differences of SUV between PET MR and PET CT were as follows: −18.8%, bone (SUV mean ); −8.0%, liver (SUV mean ); −12.2%, lung (SUV mean ); −18.1%, bone lesions (SUV mean ); −13.3%, bone lesions (SUV max ); −8.2%, soft-tissue lesions (SUV mean ); and −7.3%, soft-tissue lesions (SUV max ). The mean relative differences between PET MR and PET CTSeg were as follows: −19.0%, bone (SUV mean ); −3.5%, liver (SUV mean ); −3.3%, lung (SUV mean ); −19.3%, bone lesions (SUV mean ); −17.5%, bone lesions (SUV max ); −5.5%, soft-tissue lesions (SUV mean ); and −4.4%, soft-tissue lesions (SUV max ). The differences of SUV between PET MR and PET CT were larger than those between PET MR and PET CTSeg , in both soft tissue and soft-tissue lesions (P<0.001), but not in bone or bone lesions. In conclusion, MRAC in the newly released PET/MR system is accurate in most tissues, with SUV deviations being generally less than 10%, compared to PET/CT. In bone, however, underestimations can be substantial, which may be partially attributed to segmentation of the MR-based μ -maps.
The aim of this study was to assess the diagnostic performance of radiological imaging in differentiating xanthogranulomatous cholecystitis (XGC) from gallbladder cancer (GBC).A retrospective analysis of the radiological imaging performed in patients who had pathologically confirmed XGC or GBC between December 2004 to April 2016 was performed. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of each imaging modality, and combined imaging modalities were calculated.A total of 218 patients (XGC =109, GBC =109) were identified; 19 patients received all of abdominal ultrasound (US), contrast-enhanced ultrasound (CEUS), computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography-computed tomography (PET/CT); 21 received four of these imaging examination types; 45 received three examinations; 58 received two examinations; and 75 received only one examination. The sensitivity and specificity of CEUS was 90% and 93%, respectively, higher than abdominal US (80%, 86%), CT (71%, 92%), MRI (75%, 90%), and PET/CT (55%, 90%) (all values respective). The sensitivity, specificity, NPV, and PPV of the US combined with CEUS were 91%, 90%, 94%, and 85%, respectively. Although the specificity of CEUS + CT and CEUS + MRI were 100% and 92%, respectively, the sensitivity of CEUS + CT and CEUS + MRI were both only 67%.The Abdominal US is not sufficiently accurate to confidently guide clinical practice, and CEUS showed better diagnostic performance than the other imaging modalities in differentiating XGC from GBC. The combination of abdominal CEUS and CT is helpful for differential diagnosis, as it indicates GBC with better specificity and PPV.
Deep sternal wound infection (DSWI) is a severe complication in patients after open heart surgery (OHS). But there is a lack of appropriate imaging tool to detect the infection sites, which may lead to incomplete debridement. The present study aims to investigate the value of 18 F-fluorodeoxyglucose positron emission tomography/computed tomography (18 F-FDG PET/CT) in comparison with CT scan in diagnosing and localising DSWI. A total of 102 patients with DSWI after OHS were retrospectively collected from January 2012 to December 2017 in our hospital. All the patients had surgical debridements for DSWI with pretreatment imaging of either 18 F-FDG PET/CT or CT scan. The sensitivity, specificity, and accuracy of localising infection sites were compared between PET/CT and CT groups, with surgical, microbiological, and histopathological findings as the gold standard. The length of hospital stays and the rate of recurrence were also compared. Ten patients in the PET/CT group had a follow-up PET/CT scan after debridement, and the correlations between the changes of PET/CT findings and surgical outcomes were analysed. 18 F-FDG PET/CT is more accurate than CT in diagnosing and localising DSWI after OHS, which leads to a more successful surgical debridement with a lower rate of recurrence and a shorter length of hospital stay. In addition, follow-up PET/CT after debridement could evaluate the treatment effect.
To investigate the correlation among the maximum standardized uptake value (SUVmax) on fluorine-18-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) and tumor differentiation, size, and Ki67 in patients with moderately and poorly differentiate dintrahepatic cholangiocarcinoma (ICC).The 18F-FDG PET/CT imaging data of 116 patients with single ICC lesions confirmed by pathology were retrospectively evaluated. Pathological characteristics of the tumor such as the largest tumor diameter, differentiation, Ki67 expression, SUVmax of the primary tumor, and the tumor to normal background ratio (TNR) were recorded.Among the 116 lesions, 45, 51, and 20 lesions were classified into the moderately differentiated, moderately-poorly, and poorly differentiated groups, respectively. There were significant differences in the SUVmax (P=0.033) and TNR (P=0.044) among the three groups. Maximum SUV was significantly correlated with differentiation (r=0.244, P=0.008). When the cases were categorized according to the tumor size (group 1, ≤3cm, n=14; group 2, >3 and ≤5 cm, n=37; group 3, >5 and ≤10 cm, n=52; group 4, >10 cm, n=13), there were significant differences in the SUVmax (P<0.001) and TNR (P<0.001) among the four groups. Maximum SUV was significantly correlated with tumor size (r=0.481, P<0.001). Among the 116 lesions, 38 lesions and 78 lesions were classified into the low Ki67 and high Ki67 expression groups, respectively. There were significant differences in the SUVmax (P=0.028) and TNR (P=0.007) between the two groups. Maximum SUV was significantly correlated with Ki67 expression (r=0.242, P=0.009).In moderately and poorly differentiated ICC, the SUVmax and TNR are significantly associated with tumor differentiation, size, and Ki67 expression.
PET/MR is transferring from a powerful scientific research tool to an imaging modality in clinical routine practice. Whole body PET/MR screening usually takes 30-50 minutes to finish, during which a few factors might induce patient discomfort and further cause degraded image quality. The aim of this report is to investigate the patients' perception of the imaging procedure and its correlation with image quality.One hundred and twenty patients (63 males and 57 females, average age = 51.3 years, range 22-70 years) who had been diagnosed with cancer or had previous history of cancer were recruited and scanned with a simultaneous PET/MR system. A questionnaire was given to all patients retrospectively after the PET/MR scan, which has nine questions to assess patients' feeling of the scan on a Likert scale scoring system (1-5, 1 as most satisfied). All PET/MR images were also visually examined by two experts independently to evaluate the quality of the images. Six body locations were assessed and each location was evaluated also with a Likert scale scoring system (1-5, 5 as the best quality). Mann-Whitney U-test was used for statistical analysis to check if there is significant correlation between image quality and patient perceptions.With a total of 120 patients, 118 questionnaires were filled and returned for analysis. The patients' characteristics were summarized in Table 4. The statistics of the patients' perception in the questionnaire were illustrated in Tables 5-7. Statistical significant correlations were found between MR image quality and patients' characteristics/perception.Our results show that PET/MR scanning is generally safe and comfortable for most of the patients. Statistical analysis does not support the hypothesis that bad patient's perception leads to degraded image quality.
To evaluate the value of fluorine -18-fuoro-2-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) in evaluating synchronous multiple primary cancers (SMPC).Nineteen patients with pathologically-confirmed SMPC were collected. Clinical and 18F-FDG PET/CT characteristics of these patients were reviewed and analyzed. Maximum standardized uptake value, (SUVmax) of all lesions was measured and difference (Δ)SUVmax between the SUV of two primary tumors in each patient was calculated as: [(the larger SUVmax - the smaller SUVmax)/ the larger SUVmax]×100%.A total of 38 lesions were identified, which were most frequently located in gastrointestinal tract (n=16), followed by lung (n=10), breast (n=4), kidney (n=4), liver (n=2), pancreas (n=1) and thyroid (n=1). Pathologies of these 38 lesions were 18 adenocarcinomas, 8 squamous cell carcinomas, 4 breast invasive ductal carcinomas, 4 renal cell carcinomas, 2 hepatocellular carcinomas, 1 pancreatic ductal adenocarcinoma and 1 papillary thyroid carcinoma. The mean SUVmax of all lesions was 8.5±6.9, most of them being more than 2.5 (n=30). The mean ΔSUVmax was 57.3%±24.6%, indicating different metabolism of the primary cancers in each patient.In our center, SMPC most commonly involved the gastrointestinal tract and adenocarcinomas were the most common pathology type. 18F-FDG PET/CT was useful in the diagnosis of SMPC and the ΔSUVmax indicates different pathological origins of the synchronous cancers.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)