Dataset Information

Predictive classification of correlated targets with application to detection of metastatic cancer using functional CT imaging.

ABSTRACT: Perfusion computed tomography (CTp) is an emerging functional imaging modality that uses physiological models to quantify characteristics pertaining to the passage of fluid through blood vessels. Perfusion characteristics provide physiological correlates for neovascularization induced by tumor angiogenesis. Thus CTp offers promise as a non-invasive quantitative functional imaging tool for cancer detection, prognostication, and treatment monitoring. In this article, we develop a Bayesian probabilistic framework for simultaneous supervised classification of multivariate correlated objects using separable covariance. The classification approach is applied to discriminate between regions of liver that contain pathologically verified metastases from normal liver tissue using five perfusion characteristics. The hepatic regions tend to be highly correlated due to common vasculature. We demonstrate that simultaneous Bayesian classification yields dramatic improvements in performance in the presence of strong correlation among intra-subject units, yet remains competitive with classical methods in the presence of weak or no correlation.

SUBMITTER: Wang Y

PROVIDER: S-EPMC4575264 | biostudies-literature | 2015 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predictive classification of correlated targets with application to detection of metastatic cancer using functional CT imaging.

Wang Yuan Y Hobbs Brian P BP Hu Jianhua J Ng Chaan S CS Do Kim-Anh KA

Biometrics 20150407 3

Perfusion computed tomography (CTp) is an emerging functional imaging modality that uses physiological models to quantify characteristics pertaining to the passage of fluid through blood vessels. Perfusion characteristics provide physiological correlates for neovascularization induced by tumor angiogenesis. Thus CTp offers promise as a non-invasive quantitative functional imaging tool for cancer detection, prognostication, and treatment monitoring. In this article, we develop a Bayesian probabil ...[more]

PMID: 25851056

Similar Datasets

Project description:OBJECTIVE:To use the patient outcome endpoints overall survival and progression-free survival to evaluate functional parameters derived from dynamic contrast-enhanced CT. METHODS:69 patients with metastatic renal cell carcinoma had dynamic contrast-enhanced CT scans at baseline and after 5 and 10 weeks of treatment. Blood volume, blood flow and standardized perfusion values were calculated using deconvolution (BVdeconv, BFdeconv and SPVdeconv), blood flow and standardized perfusion values using maximum slope (BFmax and SPVmax) and blood volume and permeability surface area product using the Patlak model (BVpatlak and PS). Histogram data for each were extracted and associated to patient outcomes. Correlations and agreements were also assessed. RESULTS:The strongest associations were observed between patient outcome and medians and modes for BVdeconv, BVpatlak and BFdeconv at baseline and during the early ontreatment period (p < 0.05 for all). For the relative changes in median and mode between baseline and weeks 5 and 10, PS seemed to have opposite associations dependent on treatment. Interobserver correlations were excellent (r ≥ 0.9, p < 0.001) with good agreement for BFdeconv, BFmax, SPVdeconv and SPVmax and moderate to good (0.5 < r < 0.7, p < 0.001) for BVdeconv and BVpatlak. Medians had a better reproducibility than modes. CONCLUSION:Patient outcome was used to identify the best functional imaging parameters in patients with metastatic renal cell carcinoma. Taking patient outcome and reproducibility into account, BVdeconv, BVpatlak and BFdeconv provide the most clinically meaningful information, whereas PS seems to be treatment dependent. Standardization of acquisition protocols and post-processing software is necessary for future clinical utilization. Advances in knowledge: Taking patient outcome and reproducibility into account, BVdeconv, BVpatlak and BFdeconv provide the most clinically meaningful information. PS seems to be treatment dependent.

Project description:BackgroundBreast cancer is the most frequent cancer in women and remains the second leading cause of death in Western countries. It represents a heterogeneous group of diseases with diverse tumoral behaviour, treatment responsiveness and prognosis. While major progress in diagnosis and treatment has resulted in a decline in breast cancer-related mortality, some patients will relapse and prognosis in this cohort of patients remains poor. Treatment is determined according to tumor subtype; primarily hormone receptor status and HER2 expression. Menopausal status and site of disease relapse are also important considerations in treatment protocols.Main bodyStaging and repeated evaluation of patients with metastatic breast cancer are central to the accurate assessment of disease extent at diagnosis and during treatment; guiding ongoing clinical management. Advances have been made in the diagnostic and therapeutic fields, particularly with new targeted therapies. In parallel, oncological imaging has evolved exponentially with the development of functional and anatomical imaging techniques. Consistent, reproducible and validated methods of assessing response to therapy is critical in effectively managing patients with metastatic breast cancer.ConclusionMajor progress has been made in oncological imaging over the last few decades. Accurate disease assessment at diagnosis and during treatment is important in the management of metastatic breast cancer. CT (and BS if appropriate) is generally widely available, relatively cheap and sufficient in many cases. However, several additional imaging modalities are emerging and can be used as adjuncts, particularly in pregnancy or other diagnostically challenging cases. Nevertheless, no single imaging technique is without limitation. The authors have evaluated the vast array of imaging techniques - individual, combined parametric and multimodal - that are available or that are emerging in the management of metastatic breast cancer. This includes WB DW-MRI, CCA, novel PET breast cancer-epitope specific radiotracers and radiogenomics.

Project description:Group differences in resting state functional magnetic resonance imaging connectivity between individuals with autism and typically developing controls have been widely replicated for a small number of discrete brain regions, yet the whole-brain distribution of connectivity abnormalities in autism is not well characterized. It is also unclear whether functional connectivity is sufficiently robust to be used as a diagnostic or prognostic metric in individual patients with autism. We obtained pairwise functional connectivity measurements from a lattice of 7266 regions of interest covering the entire grey matter (26.4 million connections) in a well-characterized set of 40 male adolescents and young adults with autism and 40 age-, sex- and IQ-matched typically developing subjects. A single resting state blood oxygen level-dependent scan of 8 min was used for the classification in each subject. A leave-one-out classifier successfully distinguished autism from control subjects with 83% sensitivity and 75% specificity for a total accuracy of 79% (P = 1.1 × 10(-7)). In subjects <20 years of age, the classifier performed at 89% accuracy (P = 5.4 × 10(-7)). In a replication dataset consisting of 21 individuals from six families with both affected and unaffected siblings, the classifier performed at 71% accuracy (91% accuracy for subjects <20 years of age). Classification scores in subjects with autism were significantly correlated with the Social Responsiveness Scale (P = 0.05), verbal IQ (P = 0.02) and the Autism Diagnostic Observation Schedule-Generic's combined social and communication subscores (P = 0.05). An analysis of informative connections demonstrated that region of interest pairs with strongest correlation values were most abnormal in autism. Negatively correlated region of interest pairs showed higher correlation in autism (less anticorrelation), possibly representing weaker inhibitory connections, particularly for long connections (Euclidean distance >10 cm). Brain regions showing greatest differences included regions of the default mode network, superior parietal lobule, fusiform gyrus and anterior insula. Overall, classification accuracy was better for younger subjects, with differences between autism and control subjects diminishing after 19 years of age. Classification scores of unaffected siblings of individuals with autism were more similar to those of the control subjects than to those of the subjects with autism. These findings indicate feasibility of a functional connectivity magnetic resonance imaging diagnostic assay for autism.

Project description:PurposeCurrent standard of care conventional imaging modalities (CIM) such as X-ray computed tomography (CT) and bone scan can be limited for detection of metastatic prostate cancer and therefore improved imaging methods are an unmet clinical need. We evaluated the utility of a novel second-generation low molecular weight radiofluorinated prostate-specific membrane antigen (PSMA)-targeted positron emission tomography (PET) radiotracer, [(18)F]DCFPyL, in patients with metastatic prostate cancer.ProceduresNine patients with suspected prostate cancer recurrence, eight with CIM evidence of metastatic prostate cancer and one with biochemical recurrence, were imaged with [(18)F]DCFPyL PET/CT. Eight of the patients had contemporaneous CIM for comparison. A lesion-by-lesion comparison of the detection of suspected sites of metastatic prostate cancer was carried out between PET and CIM. Statistical analysis for estimated proportions of inter-modality agreement for detection of metastatic disease was calculated accounting for intra-patient correlation using general estimating equation (GEE) intercept-only regression models.ResultsOne hundred thirty-nine sites of PET positive [(18)F]DCFPyL uptake (138 definite, 1 equivocal) for metastatic disease were detected in the eight patients with available comparison CIM. By contrast, only 45 lesions were identified on CIM (30 definite, 15 equivocal). When lesions were negative or equivocal on CIM, it was estimated that a large portion of these lesions or 0.72 (95 % confidence interval (CI) 0.55-0.84) would be positive on [(18)F]DCFPyL PET. Conversely, of those lesions negative or equivocal on [(18)F]DCFPyL PET, it was estimated that only a very small proportion or 0.03 (95 % CI 0.01-0.07) would be positive on CIM. Delayed 2-h-post-injection time point PET yielded higher tumor radiotracer uptake and higher tumor-to-background ratios than an earlier 1-h-post-injection time point.ConclusionsA novel PSMA-targeted PET radiotracer, [(18)F]DCFPyL, was able to a large number of suspected sites of prostate cancer, many of which were occult or equivocal by CIM. This study provides strong preliminary evidence for the use of this second-generation PSMA-targeted PET radiotracer for detection of metastatic prostate cancer and lends further support for the importance of PSMA-targeted PET imaging in prostate cancer.

Dataset Information

Predictive classification of correlated targets with application to detection of metastatic cancer using functional CT imaging.

Publications

Predictive classification of correlated targets with application to detection of metastatic cancer using functional CT imaging.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets