1.
Concordance in the interpretation of PET after chemotherapy in advanced stage Hodgkin lymphoma.
Kobe, C, Kuhnert, G, Haverkamp, H, Fuchs, M, Kahraman, D, Eich, HT, Kriz, J, Baues, C, Nast-Kolb, B, Bröckelmann, PJ, et al
Nuklearmedizin. Nuclear medicine. 2015;(6):241-6
Abstract
UNLABELLED The aim was to analyze the degree of agreement between the central review panel and the local PET interpretation within the HD15 trial and its impact on subsequent treatment and progression free survival. PATIENTS, METHODS The analysis set consisted of 739 patients with residues ≥ 2.5 cm after 6 or 8 cycles of BEACOPPesc from the HD15 trial performed by the German Hodgkin Study Group. The recommendation for or against further radiotherapy was based on the central [(18)F]FDG-PET interpretation. Central PET interpretation was compared to the local PET interpretation and concordance was measured using Cohen's Kappa coefficient. Prognostic impact of the analysis of concordance between local and central PET interpretations was evaluated using progression free survival (PFS); groups were compared with the log rank test. RESULTS The central panel rated 548 of 739 patients (74%) as PET negative. Of these, 513 were also rated as PET negative in the local PET interpretation. PET positivity was seen by central reviewers in the remaining 191 patients (26%), in concordance with local reviewers in 155 cases. Even though substantial agreement was found (Cohen's Kappa 0.81), the interpretation of the central PET review panel led to a different therapeutic recommendation in 71/739 (10%) patients. PFS was equally high in groups in which the therapeutic regime had been changed on the basis of the central panel decision. CONCLUSION High concordance is found between local and central reviewers with regard to PET interpretation in residual tissue after intense chemotherapy. The existence of the central PET review panel allows the identification of additional patients as PET negative so that radiotherapy can be safely omitted (35 of 548 patients = 4.7%).
2.
Development and validation of a predictive screening tool for uninterpretable coronary CT angiography results.
Vanhecke, TE, Madder, RD, Weber, JE, Bielak, LF, Peyser, PA, Chinnaiyan, KM
Circulation. Cardiovascular imaging. 2011;(5):490-7
Abstract
BACKGROUND Coronary CT angiography (CCTA) is an excellent tool for noninvasive assessment of coronary arteries in low- to intermediate-risk individuals. However, the accuracy of CCTA heavily depends on image quality. Our objective was to develop and validate a tool to predict pre-CCTA risk of obtaining an uninterpretable result in symptomatic patients. METHODS AND RESULTS Among 8585 symptomatic patients, we identified variables independently associated with the presence of at least 1 uninterpretable major coronary segment to create the uninterpretable risk score (URS). This risk score was developed using both clinical variables and patient variables acquired at the time the CCTA was performed (heart rate and coronary calcium). The URS was then prospectively validated among an additional 915 symptomatic patients. The URS was predictive of uninterpretable results in both the development and the validation cohorts. For every 4-point increase in the URS (range, 0 to 12), the rate of at least 1 uninterpretable coronary segment per 100 CCTA studies increased ≈1.5 fold. Increased heart rate and coronary artery calcium score were predictive of uninterpretable CCTA study results. Uninterpretable results were associated with 3-month outcomes in the development cohort. CONCLUSIONS The URS can categorize patients who are likely to have at least 1 uninterpretable major coronary segment on CCTA. This may aid in appropriate patient selection for CCTA and avoiding radiation exposure in those likely to have an uninterpretable study. Clinical Trial Registration- URL: http:///www.clinicaltrials.gov. Unique identifier: NCT00640068.
3.
Hybrid retinal image registration.
Chanwimaluang, T, Fan, G, Fransen, SR
IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society. 2006;(1):129-42
Abstract
This work studies retinal image registration in the context of the National Institutes of Health (NIH) Early Treatment Diabetic Retinopathy Study (ETDRS) standard. The ETDRS imaging protocol specifies seven fields of each retina and presents three major challenges for the image registration task. First, small overlaps between adjacent fields lead to inadequate landmark points for feature-based methods. Second, the non-uniform contrast/intensity distributions due to imperfect data acquisition will deteriorate the performance of area-based techniques. Third, high-resolution images contain large homogeneous nonvascular/texureless regions that weaken the capabilities of both feature-based and area-based techniques. In this work, we propose a hybrid retinal image registration approach for ETDRS images that effectively combines both area-based and feature-based methods. Four major steps are involved. First, the vascular tree is extracted by using an efficient local entropy-based thresholding technique. Next, zeroth-order translation is estimated by maximizing mutual information based on the binary image pair (area-based). Then image quality assessment regarding the ETDRS field definition is performed based on the translation model. If the image pair is accepted, higher-order transformations will be involved. Specifically, we use two types of features, landmark points and sampling points, for affine/quadratic model estimation. Three empirical conditions are derived experimentally to control the algorithm progress, so that we can achieve the lowest registration error and the highest success rate. Simulation results on 504 pairs of ETDRS images show the effectiveness and robustness of the proposed algorithm.