Evaluation of a novel deep learning-based classifier for perifissural nodules.

OBJECTIVES: To evaluate the performance of a novel convolutional neural network (CNN) for the classification of typical perifissural nodules (PFN). METHODS: Chest CT data from two centers in the UK and The Netherlands (1668 unique nodules, 1260 individuals) were collected. Pulmonary nodules were classified into subtypes, including "typical PFNs" on-site, and were reviewed by a central clinician. The dataset was divided into a training/cross-validation set of 1557 nodules (1103 individuals) and a test set of 196 nodules (158 individuals). For the test set, three radiologically trained readers classified the nodules into three nodule categories: typical PFN, atypical PFN, and non-PFN. The consensus of the three readers was used as reference to evaluate the performance of the PFN-CNN. Typical PFNs were considered as positive results, and atypical PFNs and non-PFNs were grouped as negative results. PFN-CNN performance was evaluated using the ROC curve, confusion matrix, and Cohen's kappa. RESULTS: Internal validation yielded a mean AUC of 91.9% (95% CI 90.6-92.9) with 78.7% sensitivity and 90.4% specificity. For the test set, the reader consensus rated 45/196 (23%) of nodules as typical PFN. The classifier-reader agreement (k = 0.62-0.75) was similar to the inter-reader agreement (k = 0.64-0.79). Area under the ROC curve was 95.8% (95% CI 93.3-98.4), with a sensitivity of 95.6% (95% CI 84.9-99.5), and specificity of 88.1% (95% CI 81.8-92.8). CONCLUSION: The PFN-CNN showed excellent performance in classifying typical PFNs. Its agreement with radiologically trained readers is within the range of inter-reader agreement. Thus, the CNN-based system has potential in clinical and screening settings to rule out perifissural nodules and increase reader efficiency. KEY POINTS: • Agreement between the PFN-CNN and radiologically trained readers is within the range of inter-reader agreement. • The CNN model for the classification of typical PFNs achieved an AUC of 95.8% (95% CI 93.3-98.4) with 95.6% (95% CI 84.9-99.5) sensitivity and 88.1% (95% CI 81.8-92.8) specificity compared to the consensus of three readers.

Assessing the Accuracy of a Deep Learning Method to Risk Stratify Indeterminate Pulmonary Nodules.

Rationale: The management of indeterminate pulmonary nodules (IPNs) remains challenging, resulting in invasive procedures and delays in diagnosis and treatment. Strategies to decrease the rate of unnecessary invasive procedures and optimize surveillance regimens are needed.Objectives: To develop and validate a deep learning method to improve the management of IPNs.Methods: A Lung Cancer Prediction Convolutional Neural Network model was trained using computed tomography images of IPNs from the National Lung Screening Trial, internally validated, and externally tested on cohorts from two academic institutions.Measurements and Main Results: The areas under the receiver operating characteristic curve in the external validation cohorts were 83.5% (95% confidence interval [CI], 75.4-90.7%) and 91.9% (95% CI, 88.7-94.7%), compared with 78.1% (95% CI, 68.7-86.4%) and 81.9 (95% CI, 76.1-87.1%), respectively, for a commonly used clinical risk model for incidental nodules. Using 5% and 65% malignancy thresholds defining low- and high-risk categories, the overall net reclassifications in the validation cohorts for cancers and benign nodules compared with the Mayo model were 0.34 (Vanderbilt) and 0.30 (Oxford) as a rule-in test, and 0.33 (Vanderbilt) and 0.58 (Oxford) as a rule-out test. Compared with traditional risk prediction models, the Lung Cancer Prediction Convolutional Neural Network was associated with improved accuracy in predicting the likelihood of disease at each threshold of management and in our external validation cohorts.Conclusions: This study demonstrates that this deep learning algorithm can correctly reclassify IPNs into low- or high-risk categories in more than a third of cancers and benign nodules when compared with conventional risk models, potentially reducing the number of unnecessary invasive procedures and delays in diagnosis.

External validation of a convolutional neural network artificial intelligence tool to predict malignancy in pulmonary nodules.

BACKGROUND: Estimation of the risk of malignancy in pulmonary nodules detected by CT is central in clinical management. The use of artificial intelligence (AI) offers an opportunity to improve risk prediction. Here we compare the performance of an AI algorithm, the lung cancer prediction convolutional neural network (LCP-CNN), with that of the Brock University model, recommended in UK guidelines. METHODS: A dataset of incidentally detected pulmonary nodules measuring 5-15 mm was collected retrospectively from three UK hospitals for use in a validation study. Ground truth diagnosis for each nodule was based on histology (required for any cancer), resolution, stability or (for pulmonary lymph nodes only) expert opinion. There were 1397 nodules in 1187 patients, of which 234 nodules in 229 (19.3%) patients were cancer. Model discrimination and performance statistics at predefined score thresholds were compared between the Brock model and the LCP-CNN. RESULTS: The area under the curve for LCP-CNN was 89.6% (95% CI 87.6 to 91.5), compared with 86.8% (95% CI 84.3 to 89.1) for the Brock model (p≤0.005). Using the LCP-CNN, we found that 24.5% of nodules scored below the lowest cancer nodule score, compared with 10.9% using the Brock score. Using the predefined thresholds, we found that the LCP-CNN gave one false negative (0.4% of cancers), whereas the Brock model gave six (2.5%), while specificity statistics were similar between the two models. CONCLUSION: The LCP-CNN score has better discrimination and allows a larger proportion of benign nodules to be identified without missing cancers than the Brock model. This has the potential to substantially reduce the proportion of surveillance CT scans required and thus save significant resources.

Quantification of 18F-florbetapir PET: comparison of two analysis methods.

PURPOSE: (18)F-Florbetapir positron emission tomography (PET) can be used to image amyloid burden in the human brain. A previously developed research method has been shown to have a high test-retest reliability and good correlation between standardized uptake value ratio (SUVR) and amyloid burden at autopsy. The goal of this study was to determine how well SUVRs computed using the research method could be reproduced using an automatic quantification method, developed for clinical use. METHODS: Two methods for the quantitative analysis of (18)F-florbetapir PET were compared in a diverse clinical population of 604 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and in a group of 74 younger healthy controls (YHC). Cortex to cerebellum SUVRs were calculated using the research method, which is based on SPM, yielding 'research SUVRs', and using syngo.PET Amyloid Plaque, yielding 'sPAP SUVRs'. RESULTS: Mean cortical SUVRs calculated using the two methods for the 678 subjects were correlated (r = 0.99). Linear regression of sPAP SUVRs on research SUVRs was used to convert the research method SUVR threshold for florbetapir positivity of 1.10 to a corresponding threshold of 1.12 for sPAP. Using the corresponding thresholds, categorization of SUVR values were in agreement between research and sPAP SUVRs for 96.3 % of the ADNI images. SUVRs for all YHC were below the corresponding thresholds. CONCLUSION: Automatic florbetapir PET quantification using sPAP yielded cortex to cerebellum SUVRs which were correlated and in good agreement with the well-established research method. The research SUVR threshold for florbetapir positivity was reliably converted to a corresponding threshold for sPAP SUVRs.

Dependency of cardiac rubidium-82 imaging quantitative measures on age, gender, vascular territory, and software in a cardiovascular normal population.

OBJECTIVES: Recent technological improvements to PET imaging equipment combined with the availability of software optimized to calculate regional myocardial blood flow (MBF) and myocardial flow reserve (MFR) create a paradigm shifting opportunity to provide new clinically relevant quantitative information to cardiologists. However, clinical interpretation of the MBF and MFR is entirely dependent upon knowledge of MBF and MFR values in normal populations and subpopulations. This work reports Rb-82-based MBF and MFR measurements for a series of 49 verified cardiovascularly normal subjects as a preliminary baseline for future clinical studies. METHODS: Forty-nine subjects (24F/25M, ages 41-69) with low probability for coronary artery disease and with normal exercise stress test were included. These subjects underwent rest/dipyridamole stress Rb-82 myocardial perfusion imaging using standard clinical techniques (40 mCi injection, 6-minute acquisition) using a Siemens Biograph 40 PET/CT scanner with high count rate detector option. List mode data was rehistogrammed into 26 dynamic frames (12 × 5 seconds, 6 × 10 seconds, 4 × 20 seconds, 4 × 40 seconds). Cardiac images were processed, and MBF and MFR calculated using Siemens syngo MBF, PMOD, and FlowQuant software using a single compartment Rb-82 model. RESULTS: Global myocardial blood flow under pharmacological stress for the 24 females as measured by PMOD, syngo MBF, and FlowQuant were 3.10 ± 0.72, 2.80 ± 0.66, and 2.60 ± 0.63 mL·minute(-1)·g(-1), and for the 25 males was 2.60 ± 0.84, 2.33 ± 0.75, 2.15 ± 0.62 mL·minute(-1)·g(-1), respectively. Rest flows for PMOD, syngo MBF, and FlowQuant averaged 1.32 ± 0.42, 1.20 ± 0.33, and 1.06 ± 0.38 mL·minute(-1)·g(-1) for the female subjects, and 1.12 ± 0.29, 0.90 ± 0.26, and 0.85 ± 0.24 mL·minute(-1)·g(-1) for the males. Myocardial flow reserves for PMOD, syngo MBF, and FlowQuant for the female normals were calculated to be 2.50 ± 0.80, 2.53 ± 0.67, 2.71 ± 0.90, and 2.50 ± 1.19, 2.85 ± 1.19, 2.94 ± 1.31 mL·minute(-1)·g(-1) for males. CONCLUSION: Quantitative normal MBF and MFR values averaged for age and sex have been compiled for three commercial pharmacokinetic software packages. The current collection of data consisting of 49 subjects resulted in several statistically significant conclusions that support the need for a software specific, age, and sex-matched database to aid in interpretation of quantitative clinical myocardial perfusion studies.

A fluorous and click approach for screening potential PET probes: Evaluation of potential hypoxia biomarkers.

Radiopharmaceuticals for nuclear imaging are essentially targeting molecules, labeled with short-lived radionuclides (e.g., F-18 for PET). A significant drawback of radiopharmaceuticals development is the difficulty to access radiolabeled molecule libraries for initial in vitro evaluation, as radiolabeling has to be optimized for each individual molecule. The present paper discloses a method for preparing libraries of (18)F-labeled radiopharmaceuticals using both the fluorous-based (18)F-radiochemistry and the Huisgen 1,3-dipolar (click) conjugation reaction. As a proof of concept, this approach allowed us to obtain a series of readily accessible (18)F-radiolabeled nitroaromatic molecules, for exploring their structure-activity relationship and further in vitro evaluation of their hypoxic selectivity.

Novel Fast Marching for Automated Segmentation of the Hippocampus (FMASH): method and validation on clinical data.

With hippocampal atrophy both a clinical biomarker for early Alzheimer's Disease (AD) and implicated in many other neurological and psychiatric diseases, there is much interest in the accurate, reproducible delineation of this region of interest (ROI) in structural MR images. Here we present Fast Marching for Automated Segmentation of the Hippocampus (FMASH): a novel approach using the Sethian Fast Marching (FM) technique to grow a hippocampal ROI from an automatically-defined seed point. Segmentation performance is assessed on two separate clinical datasets, utilising expert manual labels as gold standard to quantify Dice coefficients, false positive rates (FPR) and false negative rates (FNR). The first clinical dataset (denoted CMA) contains normal controls (NC) and atrophied AD patients, whilst the second is a collection of NC and bipolar (BP) patients (denoted BPSA). An optimal and robust stopping criterion is established for the propagating FM front and the final FMASH segmentation estimates compared to two commonly-used methods: FIRST/FSL and Freesurfer (FS). Results show that FMASH outperforms both FIRST and FS on the BPSA data, with significantly higher Dice coefficients (0.80±0.01) and lower FPR. Despite some intrinsic bias for FIRST and FS on the CMA data, due to their training, FMASH performs comparably well on the CMA data, with an average bilateral Dice coefficient of 0.82±0.01. Furthermore, FMASH most accurately captures the hippocampal volume difference between NC and AD, and provides a more accurate estimation of the problematic hippocampus-amygdala border on both clinical datasets. The consistency in performance across the two datasets suggests that FMASH is applicable to a range of clinical data with differing image quality and demographics.

The traceless Staudinger ligation for indirect 18F-radiolabelling.

The Staudinger ligation of phosphine-substituted thioesters with (18)F-fluoroethylazide has been successfully applied to access (18)F-labelled molecules in radiochemical yields superior to 95%; the first fluorous variant of a Staudinger radio-ligation has been validated.

Lung cancer prediction by Deep Learning to identify benign lung nodules.

INTRODUCTION: Deep Learning has been proposed as promising tool to classify malignant nodules. Our aim was to retrospectively validate our Lung Cancer Prediction Convolutional Neural Network (LCP-CNN), which was trained on US screening data, on an independent dataset of indeterminate nodules in an European multicentre trial, to rule out benign nodules maintaining a high lung cancer sensitivity. METHODS: The LCP-CNN has been trained to generate a malignancy score for each nodule using CT data from the U.S. National Lung Screening Trial (NLST), and validated on CT scans containing 2106 nodules (205 lung cancers) detected in patients from from the Early Lung Cancer Diagnosis Using Artificial Intelligence and Big Data (LUCINDA) study, recruited from three tertiary referral centers in the UK, Germany and Netherlands. We pre-defined a benign nodule rule-out test, to identify benign nodules whilst maintaining a high sensitivity, by calculating thresholds on the malignancy score that achieve at least 99 % sensitivity on the NLST data. Overall performance per validation site was evaluated using Area-Under-the-ROC-Curve analysis (AUC). RESULTS: The overall AUC across the European centers was 94.5 % (95 %CI 92.6-96.1). With a high sensitivity of 99.0 %, malignancy could be ruled out in 22.1 % of the nodules, enabling 18.5 % of the patients to avoid follow-up scans. The two false-negative results both represented small typical carcinoids. CONCLUSION: The LCP-CNN, trained on participants with lung nodules from the US NLST dataset, showed excellent performance on identification of benign lung nodules in a multi-center external dataset, ruling out malignancy with high accuracy in about one fifth of the patients with 5-15 mm nodules.

Use of the Resection Map system as guidance during hepatectomy.

BACKGROUND: The objective of this work is to evaluate a new concept of intraoperative three-dimensional (3D) visualization system to support hepatectomy. The Resection Map aims to provide accurate cartography for surgeons, who can therefore anticipate risks, increase their confidence and achieve safer liver resection. METHODS: In an experimental prospective cohort study, ten consecutive patients admitted for hepatectomy to three European hospitals were selected. Liver structures (portal veins, hepatic veins, tumours and parenchyma) were segmented from a recent computed tomography (CT) study of each patient. The surgeon planned the resection preoperatively and read the Resection Map as reference guidance during the procedure. Objective (amount of bleeding, tumour resection margin and operating time) and subjective parameters were retrieved after each case. RESULTS: Three different surgeons operated on seven patients with the navigation aid of the Resection Map. Veins displayed in the Resection Map were identified during the surgical procedure in 70.1% of cases, depending mainly on size. Surgeons were able to track resection progress and experienced improved orientation and increased confidence during the procedure. CONCLUSIONS: The Resection Map is a pragmatic solution to enhance the orientation and confidence of the surgeon. Further studies are needed to demonstrate improvement in patient safety.

Positron emission tomography PET/CT harmonisation study of different clinical PET/CT scanners using commercially available software.

OBJECTIVES: Harmonisation is the process whereby standardised uptake values from different scanners can be made comparable. This PET/CT pilot study aimed to evaluate the effectiveness of harmonisation of a modern scanner with image reconstruction incorporating resolution recovery (RR) with another vendor older scanner operated in two-dimensional (2D) mode, and for both against a European standard (EARL). The vendor-proprietary software EQ•PET was used, which achieves harmonisation with a Gaussian smoothing. A substudy investigated effect of RR on harmonisation. METHODS: Phantom studies on each scanner were performed to optimise the smoothing parameters required to achieve successful harmonisation. 80 patients were retrospectively selected; half were imaged on each scanner. As proof of principle, a cohort of 10 patients was selected from the modern scanner subjects to study the effects of RR on harmonisation. RESULTS: Before harmonisation, the modern scanner without RR adhered to EARL specification. Using the phantom data, filters were derived for optimal harmonisation between scanners and with and without RR as applicable, to the EARL standard. The 80-patient cohort did not reveal any statistically significant differences. In the 10-patient cohort SUVmax for RR > no RR irrespective of harmonisation but differences lacked statistical significance (one-way ANOVA F(3.36) = 0.37, p = 0.78). Bland-Altman analysis showed that harmonisation reduced the SUVmax ratio between RR and no RR to 1.07 (95% CI 0.96-1.18) with no outliers. CONCLUSIONS: EQ•PET successfully enabled harmonisation between modern and older scanners and against the EARL standard. Harmonisation reduces SUVmax and dependence on the use of RR in the modern scanner. ADVANCES IN KNOWLEDGE: EQ•PET is feasible to harmonise different PET/CT scanners and reduces the effect of RR on SUVmax.

Using EQ·PET to reduce reconstruction-dependent variations in [18F]FDG-PET brain imaging.

This study aims at assessing whether EANM harmonisation strategy combined with EQ·PET methodology could be successfully applied to harmonize brain 2-deoxy-2[18F]fluoro-D-glucose ([18F]FDG) positron emission tomography (PET) images. The NEMA NU 2 body phantom was prepared according to the EANM guidelines with an [18F]FDG solution. Raw PET phantom data were reconstructed with three different reconstruction protocols frequently used in clinical PET brain imaging: ([Formula: see text]) Ordered subset expectation maximization (OSEM) 3D with time of flight (TOF), 2 iterations and 21 subsets; ([Formula: see text]) OSEM 3D with TOF, 6 iterations and 21 subsets; and ([Formula: see text]) OSEM 3D with TOF, point spread function (PSF), and 8 iterations and 21 subsets. EQ·PET filters were computed as the Gaussian smoothing that best independently aligned the recovery coefficients (RCs) of reconstructions [Formula: see text] and [Formula: see text] with the RCs of the reference reconstruction, [Formula: see text]. The performance of the EQ·PET filter to reduce variations in quantification due to differences in reconstruction was investigated using clinical PET brain images of 35 early-onset Alzheimer's disease (EOAD) patients. Qualitative assessments and multiple quantitative metrics on the cortical surface at different scale levels with or without partial volume effect correction were evaluated on the [18F]FDG brain data before and after application of the EQ·PET filter. The EQ·PET methodology succeeded in finding the optimal smoothing that minimised root-mean-square error (RMSE) calculated using human brain [18F]FDG-PET datasets of EOAD patients, providing harmonized comparisons in the neurological context. Performance was superior for TOF than for TOF + PSF reconstructions. Results showed the capability of the EQ·PET methodology to minimize reconstruction-induced variabilities between brain [18F]FDG-PET images. However, moderate variabilities remained after harmonizing PSF reconstructions with standard non-PSF OSEM reconstructions, suggesting that precautions should be taken when using PSF modelling.

Assessing Reliability of Myocardial Blood Flow After Motion Correction With Dynamic PET Using a Bayesian Framework.

The estimation of myocardial blood flow (MBF) in dynamic PET can be biased by many different processes. A major source of error, particularly in clinical applications, is patient motion. Patient motion, or gross motion, creates displacements between different PET frames as well as between the PET frames and the CT-derived attenuation map, leading to errors in MBF calculation from voxel time series. Motion correction techniques are challenging to evaluate quantitatively and the impact on MBF reliability is not fully understood. Most metrics, such as signal-to-noise ratio (SNR), are characteristic of static images, and are not specific to motion correction in dynamic data. This study presents a new approach of estimating motion correction quality in dynamic cardiac PET imaging. It relies on calculating a MBF surrogate, K1 , along with the uncertainty on the parameter. This technique exploits a Bayesian framework, representing the kinetic parameters as a probability distribution, from which the uncertainty measures can be extracted. If the uncertainty extracted is high, the parameter studied is considered to have high variability - or low confidence - and vice versa. The robustness of the framework is evaluated on simulated time activity curves to ensure that the uncertainties are consistently estimated at the multiple levels of noise. Our framework is applied on 40 patient datasets, divided in 4 motion magnitude categories. Experienced observers manually realigned clinical datasets with 3D translations to correct for motion. K1 uncertainties were compared before and after correction. A reduction of uncertainty after motion correction of up to 60% demonstrates the benefit of motion correction in dynamic PET and as well as provides evidence of the usefulness of the new method presented.

Cardiac Displacement During 13N-Ammonia Myocardial Perfusion PET/CT: Comparison Between Adenosine- and Regadenoson-Induced Stress.

This study investigated differences in cardiac displacement during adenosine stress versus regadenoson stress in 13N-ammonia (13NH3) MP PET/CT scans. Methods: In total, 61 myocardial perfusion PET/CT scans were acquired using either adenosine (n = 30) or regadenoson (n = 31) as a stressor. For both groups, cardiac displacement during rest and stress was measured 3-dimensionally, relative to either a fixed reference frame or the previous frame, in each 1-min frame of a list-mode PET acquisition of 25 min. All stress scans were additionally evaluated for the presence of motion artifacts. Also, the tolerability of the agents and the occurrence of side effects were compared between groups. Results: Significantly larger cardiac displacement during stress was detected in the adenosine group than in the regadenoson group, reflected by both maximal cardiac displacement (P = 0.022) and mean cardiac displacement (P = 0.001). The duration of the movement was typically shorter in the regadenoson group. Frames with cardiac displacement of at least 5 mm were observed nearly twice as frequently when adenosine was used instead of regadenoson. Conclusion: The displacement during regadenoson stress is of lower amplitude and shorter duration than that during adenosine stress and may therefore contribute to a lower incidence of motion artifacts on PET/CT scans.

Development and validation of clinical prediction models to risk stratify patients presenting with small pulmonary nodules: a research protocol.

INTRODUCTION: Lung cancer is a common cancer, with over 1.3 million cases worldwide each year. Early diagnosis using computed tomography (CT) screening has been shown to reduce mortality but also detect non-malignant nodules that require follow-up scanning or alternative methods of investigation. Practical and accurate tools that can predict the probability that a lung nodule is benign or malignant will help reduce costs and the risk of morbidity and mortality associated with lung cancer. METHODS: Retrospectively collected data from 1500 patients with pulmonary nodule(s) of up to 15 mm detected on routinely performed CT chest scans aged 18 years old or older from three academic centres in the UK will be used to to develop risk stratification models. Radiological, clinical and patient characteristics will be combined in multivariable logistic regression models to predict nodule malignancy. Data from over 1000 participants recruited in a prospective phase of the study will be used to evaluate model performance. Discrimination, calibration and clinical utility measures will be presented.

Classification of amyloid status using machine learning with histograms of oriented 3D gradients.

Brain amyloid burden may be quantitatively assessed from positron emission tomography imaging using standardised uptake value ratios. Using these ratios as an adjunct to visual image assessment has been shown to improve inter-reader reliability, however, the amyloid positivity threshold is dependent on the tracer and specific image regions used to calculate the uptake ratio. To address this problem, we propose a machine learning approach to amyloid status classification, which is independent of tracer and does not require a specific set of regions of interest. Our method extracts feature vectors from amyloid images, which are based on histograms of oriented three-dimensional gradients. We optimised our method on 133 18F-florbetapir brain volumes, and applied it to a separate test set of 131 volumes. Using the same parameter settings, we then applied our method to 209 11C-PiB images and 128 18F-florbetaben images. We compared our method to classification results achieved using two other methods: standardised uptake value ratios and a machine learning method based on voxel intensities. Our method resulted in the largest mean distances between the subjects and the classification boundary, suggesting that it is less likely to make low-confidence classification decisions. Moreover, our method obtained the highest classification accuracy for all three tracers, and consistently achieved above 96% accuracy.

Semiquantitative slab view display for visual evaluation of 123I-FP-CIT SPECT.

OBJECTIVE: Dopamine transporter single-photon emission computed tomography (SPECT) with I-FP-CIT is used widely in the diagnosis of clinically uncertain parkinsonian syndromes. In terms of the evaluation of FP-CIT SPECT, some practice guidelines state that visual interpretation alone is generally sufficient in clinical patient care, whereas other guidelines consider semiquantitative analysis of striatal dopamine transporter availability mandatory. This discrepancy might be because of a relative lack of widely available display tools for FP-CIT SPECT. In this study, we evaluate a semiquantitative slab view display optimized for visual evaluation of FP-CIT SPECT that might resolve the discrepancy. PATIENTS AND METHODS: The reconstructed FP-CIT SPECT image was stereotactically normalized and scaled voxel by voxel to the mean uptake in the entire brain without striata. From the resulting distribution volume ratio image, a 12-mm-thick transversal slice (slab) through the striata was displayed with a standard colour table with predefined fixed thresholds on the distribution volume ratio. Visual scoring of the semiquantitative slab view was performed twice by four independent readers in 235 unselected patients. The specific binding ratio in the caudate and putamen was computed by fully automated semiquantitative analysis with predefined standard regions of interest in template space. RESULTS: Intrarater and inter-rater agreement of binary visual categorization as 'normal' or 'reduced' was excellent (mean Cohen's κ=0.88 and 0.83, respectively). The area under the receiver-operator characteristic curve of the specific putamen-binding ratio for differentiation between visually normal and visually reduced (majority read) was 0.96. CONCLUSION: Visual interpretation of FP-CIT SPECT on the basis of the semiquantitative slab view display provides excellent stability within and between readers as well as very high agreement with semiquantitative analysis. This suggests that the slab view display enables reliable visual interpretation of FP-CIT SPECT in clinical routine patient care.

Improving influx constant and ratio estimation in FDOPA brain PET analysis for Parkinson's disease.

UNLABELLED: PET studies using l-3,4-dihydroxy-6-(18)F-fluorophenylalanine have been applied to assess the diminished functionality of the striatum in patients with suspected Parkinson's disease. Two techniques for analyzing such studies are ratio methods and graphically computed influx constants. We propose a method for improving the consistency with which scans obtained by either of these techniques are analyzed. The method is based on a fully 3-dimensional analysis of the images. METHODS: Fifty-one dynamically acquired datasets were corrected for motion before analysis. Regions of interest (ROIs) for the analysis were determined by manual affine registration to a standard template, using a separate transformation for each ROI. Indicator values for each ROI were computed by averaging the values of voxels having the highest activity within a specified proportion of the voxels in the ROI, to increase the robustness to perturbations in the ROI position. Sensitivity was analyzed by examining the variation in results obtained when the ROIs were translated by up to 6 mm. RESULTS: We observed significant percentage differences in the computed influx constant before and after motion correction (mean variation +/- SD, -0.75% +/- 9.5% averaged over all regions of all patients). Our method was robust to placement of the cerebellum ROI, whereas a 2- dimensional analysis based on hand-drawn ROIs was associated with a 2- to 3-fold greater percentage variation in uptake for translations of 2 mm or more in ROI position. When we compared the 2 quantification techniques, our influx constants and ratios correlated at r(2) = 0.91, P < 0.0001. CONCLUSION: Motion correction is an important step for computing reliable results in dynamic studies. The robustness of the results can be increased further by using standard normalized volumetric ROIs and by using the average value of a specified proportion of the voxels with the highest activity in the ROI as an indicator for that ROI. Influx constant values computed using our analysis technique closely correlated to values computed with ratio methods using this general approach.

Patient-specific biomechanical model for the prediction of lung motion from 4-D CT images.

This paper presents an approach to predict the deformation of the lungs and surrounding organs during respiration. The framework incorporates a computational model of the respiratory system, which comprises an anatomical model extracted from computed tomography (CT) images at end-expiration (EE), and a biomechanical model of the respiratory physiology, including the material behavior and interactions between organs. A personalization step is performed to automatically estimate patient-specific thoracic pressure, which drives the biomechanical model. The zone-wise pressure values are obtained by using a trust-region optimizer, where the estimated motion is compared to CT images at end-inspiration (EI). A detailed convergence analysis in terms of mesh resolution, time stepping and number of pressure zones on the surface of the thoracic cavity is carried out. The method is then tested on five public datasets. Results show that the model is able to predict the respiratory motion with an average landmark error of 3.40 ±1.0 mm over the entire respiratory cycle. The estimated 3-D lung motion may constitute as an advanced 3-D surrogate for more accurate medical image reconstruction and patient respiratory analysis.

Automated 3-D echocardiography analysis compared with manual delineations and SPECT MUGA.

A major barrier for using 3-D echocardiography for quantitative analysis of heart function in routine clinical practice is the absence of accurate and robust segmentation and tracking methods necessary to make the analysis automatic. In this paper, we present an automated three-dimensional (3-D) echocardiographic acquisition and image-processing methodology for assessment of left ventricular (LV) function. We combine global image information provided by a novel multiscale fuzzy-clustering segmentation algorithm, with local boundaries obtained with phase-based acoustic feature detection. We then use the segmentation results to fit and track the LV endocardial surface using a 3-D continuous transformation. To our knowledge, this is the first report of a completely automated method. The protocol is evaluated in a small clinical case study (nine patients). We compare ejection fractions (EFs) computed with the new approach to those obtained using the standard clinical technique, single-photon emission computed tomography multigated acquisition. Errors on six datasets were found to be within six percentage points. A further two, with poor image quality, improved upon EFs from manually delineated contours, and the last failed due to artifacts in the data. Volume-time curves were derived and the results compared to those from manual segmentation. Improvement over an earlier published version of the method is noted.