Atlas of Clinical Positron Emission Tomography 2nd Edition

Free download. Book file PDF easily for everyone and every device. You can download and read online Atlas of Clinical Positron Emission Tomography 2nd Edition file PDF Book only if you are registered here. And also you can download or read online all Book PDF file that related with Atlas of Clinical Positron Emission Tomography 2nd Edition book. Happy reading Atlas of Clinical Positron Emission Tomography 2nd Edition Bookeveryone. Download file Free Book PDF Atlas of Clinical Positron Emission Tomography 2nd Edition at Complete PDF Library. This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats. Here is The CompletePDF Book Library. It's free to register here to get Book file PDF Atlas of Clinical Positron Emission Tomography 2nd Edition Pocket Guide.

Please take this quick survey to tell us about what happens after you publish a paper. Annals of Nuclear Medicine. We compared the diagnostic capacity of the dynamic and DTP scans using receiver operating characteristic ROC analyses. In the visual analyses of LN metastases, a sensitivity of The sensitivity of the 1-h early and 2-h delayed scans was The visual analysis of the dynamic second phase had the highest accuracy. We have no conflicts of interest with regard to this paper. This prospective study was approved by our institutional review board and ethics committee in Tokushima University Hospital.

Skip to main content. Advertisement Hide. Original Article First Online: 25 March Results In the visual analyses of LN metastases, a sensitivity of This is a preview of subscription content, log in to check access. Introduction Esophageal cancer is the seventh most common cancer in Taiwanese males [ 1 ], and the sixth most frequently diagnosed cancer worldwide [ 2 ]. Conventional workup The conventional workup CWU included computed tomography CT , barium esophagram, and panendoscopy.

Treatment All of the treatment decisions were taken after discussion in the tumor board. Statistical analysis Descriptive statistics were summarized using percentiles, means, medians, and standard deviations. Results Patient population Of the patients without distant metastases, had a diagnosis of SCC Download: PPT. Metachronous cancers At the end of the follow-up, of the study patients were alive. Discussion In our study conducted among patients with potentially curable esophageal SCC, we identified 17 cases 4. Figure 2. Figure 3. Distribution of synchronous tumors in patients with esophageal cancer according to the published studies.

Conclusions The most common form of esophageal cancer in Taiwan is SCC, and the most frequent synchronous lesions in this patient group are head and neck malignancies. Acknowledgments The authors express their gratitude to the members of the Cancer Center, Chang Gung Memorial Hospital, for their kind support. References 1. Accessed 9 August Bethesda, MD.

Accessed 9 Auguest Ann Thorac Surg PubMed: View Article Google Scholar 5. Surg Today View Article Google Scholar 6. Jpn J Clin Oncol View Article Google Scholar 7.

World J Surg View Article Google Scholar 8. J Nucl Med View Article Google Scholar 9. J Gastrointest Surg View Article Google Scholar Clin Nucl Med NCCN clinicsl practice guidelines in oncology - Esophageal and esophagogastric junction cancers. Version 2. National Comprehensive Cancer Network. Accessed 6 July Effects of chronology, geography and ethnicity.

  • A whole-body FDG PET/MR atlas for multiparametric voxel-based analysis | Scientific Reports.
  • Positron Emission Tomography in the Routine Management of Coronary Artery Disease?
  • Quantity and Quality in Social Research (Contemporary Social Research)?

J Gastroenterol Hepatol J Clin Oncol J Epidemiol Available online at: doi JE Eur J Clin Invest Cancer Epidemiol Biomarkers Prev J Hum Genet Carcinogenesis World J Gastroenterol Seitz HK, Stickel F Acetaldehyde as an underestimated risk factor for cancer development: role of genetics in ethanol metabolism. Genes Nutr 5: Eur Radiol However, a quantitative bias was seen for adipose tissue, for which the male and female fat fraction and male SUV measurements showed statistically significant differences between native and reference space.

Fat fraction, organ volume and SUV mean measured from manual segmentations for the male and female atlas subjects. For the fat fraction and SUV mean , measurements have been made in reference atlas and native space. The mean is shown with the standard deviation as error bars. Statistical significance is indicated with an asterisk.

A representative coronal slice of the registered water fraction and FDG imaging data, and automatically detected and manually segmented lesions for the male and female subjects are shown in Fig. A total of 42 lesions were outlined manually; 38 for the female and 4 for the male. The same total number of lesions was detected automatically, but 31 for the female and 11 for the male. The total number of false positives and false negatives were 21 and 11 lesions, respectively. The false positives mainly stem from physiological uptake in the brain, bladder, heart and intestines, and the false negatives mainly stem from small nodules.

One false negative however corresponded to a larger lesion in the thoracic vertebrae.


FDG anomaly detection task performed for the male left and female right subjects with suspected oesophageal cancer and lymphoma, respectively. For each subject, a representative coronal slice is shown of the registered water fraction and normalised FDG uptake images, and of the automatically detected and manually segmented tumours.

Detected and segmented lesions are displayed in red and overlaid on the water fraction image. A false positive lesion can be seen for the male subject, likely due to increased motion artefacts in the heart. After applying the lesion size and FDG uptake restrictions to the manual segmentation, 20 lesions remained 18 and 2 for the female and male subject, respectively.

The automatic method however failed to separate lesions on five occasions, for which two manually outlined close-lying lesions were detected as one. Good agreement was obtained when comparing SUV max for corresponding manual and automatically defined lesions Fig. Correlation for SUV max for corresponding tumours detected automatically and outlined manually. The automatic method failed to separate close-lying manually outlined lesions on five occasions.

The measured characteristics for these manual lesions were combined, meaning characteristics from 14 corresponding tumours in two subjects are plotted. Transaxial slices of the atlas and test subject fat fraction data, together with the results of the liver fat detection task, are shown in Fig. The detection task produced a probability map with p-values ranging from 0 to 0. The test detected abnormal fat infiltration in the liver. Fat fraction anomaly detection task performed for a male subject with an elevated liver fat infiltration.

Register for a free account

Representative transaxial slices of the single subject fat fraction data, atlas fat fraction data and automatically detected fat infiltration. All images are in reference space. Using whole-body image registration, an atlas containing quantitative parametric maps of fat and water fraction MR, local tissue volume and FDG PET was created, with separate entries for male and female subjects.

  • Brain Protection: Morphological, Pathophysiological and Clinical Aspects.
  • The Outback Bridal Rescue (Outback Knights) (Harlequin Presents # 2427).
  • WAHL - Hardcover / Medicine / Medicine & Nursing: Books?

This study expands on a previously published study in which the registration method was described in detail for fat-water MR data acquired on a 1. In the current study, an improved registration method was utilised 19 and it was shown that the concept of a whole-body atlas can be moved to a 3. Furthermore, quantitative measurements on fat content, SUV and tissue volume were in this study performed to assess the adequacy of the content of the atlas.

Specific aspects of incorporating PET into the atlas have also been described in this study, including intensity normalisation and bias assessment. As proof of concept, voxel-wise statistical comparisons were performed between the atlas and i two subjects with suspected malignant disease and ii one subject with elevated liver fat infiltration. The image registration procedure proved robust for the majority of subjects included in the atlas. Problems were however encountered, with three subjects excluded due to poor image registration results in the abdominal area. In general, this is a challenging region for the registration method due to the large physiological variability seen in this part of the body.

It was also noted that standardised subject positioning is paramount for good registration results, seen both as registration failures occurred for two subjects and in the increased variability for the feet in the atlas. The registration accuracy could potentially also be affected by the choice of reference subject, ideally chosen to not bias the final atlas.

In this study male and female reference subjects were chosen based on BMI and visual assessment of image quality. An alternative method could be to use group wise registration 24 , as such avoiding the uncertainties associated with choosing a single reference subject. This was necessary as the artefacts both impact the image registration accuracy and also affect the MR intensity values.

The latter also has an effect on the attenuation correction values used in PET image reconstruction and subsequently the PET intensity values. MR images containing metal implant artefacts can however be registered to and statistically compared with the atlas. Depending on the severity of the metal artefact, this could however affect the registration accuracy locally or globally. In addition, when performing voxel-wise statistical analysis, the area of metal artefact will most likely show a statistically significant change compared to the atlas.

Further studies on the impact of metal artefacts on both the registration accuracy and statistical analysis are needed to fully understand the effect on automated analysis. Manual tissue segmentations were performed in reference and native space and average fat fraction, organ volume and SUV mean were measured for atlas subjects Fig.

The amount of muscle fat infiltration is also comparable to previous studies, but published numbers vary due to both sequence parameters 26 and choice of muscle group The mean organ volumes obtained for the atlas subjects overlap with previous publications 28 , 29 , 30 , but there is a tendency for smaller volumes measured in particular for the female cohort.

Related Articles

Differences in the imaging protocols and segmentation methods have been shown to affect organ volumetry 31 , which could be an explanation to the discrepancy. As this study utilises the Jacobian determinant data to calculate organ volume, the registration accuracy also affects the measurements. A further reason for the somewhat smaller organ volumes could be differences in breathing instructions In this study, shallow breathing was utilised during image acquisition.

An organ size difference between genders was observed, with the male cohort exhibiting larger mean lungs, liver and spleen volumes compared to the female cohort. SUV mean corresponded well with previously published data No quantitative bias was seen for the majority of tissues cerebellum, lungs, aorta, liver, spleen and muscle when comparing measurements performed in native and reference space. A likely explanation for this is that an abnormal stretching of fat is occasionally obtained in the registration process, when the subject data being registered includes an area of very low fat volume and the corresponding area in the reference person includes a higher fat volume.

The automated tumour detection showed a strong correlation for SUV max compared with manual outlining. It was however seen that SUV max does change when measured in the registered data set Fig. To prevent this, the atlas could instead be deformed to native space and measurements performed on the original patient data.

Another option which should be addressed in the future is to scale the image intensity values to preserve the total counts in the image by using the Jacobian determinant data, often referred to as modulation. In the current implementation this is not performed, and instead the count concentration is preserved. Usage of this feature is most likely application-dependent as has been shown in brain imaging studies 33 , Future work also includes comparing the automated method with semi-automatic segmentation methods commonly used for PET imaging data A problem that was noted with the anomaly detection, that also needs addressing, is that normal physiological variations can be detected as non-normal.

This was seen as false-positive lesions for e. For the anomaly detection on fat infiltration, the probability map highlighted the liver as significantly different compared to the atlas. As a proof of concept, the utility of the fat-water MR and FDG parts of the atlas was included in this study. However, the local tissue volume component of the atlas can be used in the same manner and has potential usage areas in e.

Systemic diseases benefit the most from the whole-body approach that Imiomics provides. The voxel-by-voxel correspondence that is achieved makes group-wise analysis and correlation studies potential usage areas, as well as longitudinal studies of the same subject scanned over time. Furthermore, it is possible to create organ and tissue segmentations in the atlas, which after registration can be applied to individual subjects. This could be used as an aid to speed up the usually time-consuming manual outlining procedure. This study has some limitations. Conservative inclusion criteria were used when creating the atlas, with 51 subjects excluded due to suspected pathology, poor image quality or protocol non-adherence.

Atlas of Clinical Positron Emission Tomography 2nd Edition - Sally Barrington - Google книги

This is necessary in order to create a healthy control atlas, but it has the consequence of reducing the power of the atlas. A larger cohort would be needed to counter this issue. The atlas contains a limited age span 51—76 years and BMI range Hence, the atlas cannot be used for paediatric cases and care should be taken for obese and younger adults. For cancer applications, however, the age span included is highly relevant. An inherent limitation of the data acquisition is the difference in acquisition time between the PET and MR data, in general giving rise to more motion artefacts in the PET data and as such spatial differences between the PET and MR data sets, in particular adjacent to moving organs.

Further investigations and improvements are needed with regards to optimising the PET intensity normalisation e. As a proof of concept, two anomaly detection tasks were performed using the atlas. The first task successfully detected non-normal FDG uptake for two subjects with suspected tumour disease. The second task automatically found abnormalities in liver fat infiltration. This proof of concept has thus demonstrated that automated whole-body voxel-wise analysis is possible for these multiparametric datasets.

The datasets generated and analysed during the current study are available from the corresponding author on reasonable request. Padhani, A. Multiparametric imaging of tumor response to therapy.

Radiology , — Johansson, E. Bailey, D. Imaging Biol. Eisenhauer, E. Cancer Oxf.

Aerts, H. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Wahl, R. Strand, R.