Sabina Tangaro

Comparing and combining algorithms for computer-aided detection of pulmonary nodules in computed tomography scans: The ANODE09 study

Numerous publications and commercial systems are available that deal with automatic detection of pulmonary nodules in thoracic computed tomography scans, but a comparative study where many systems are applied to the same data set has not yet been performed. This paper introduces ANODE09 ( http://anode09.isi.uu.nl), a database of 55 scans from a lung cancer screening program and a web-based framework for objective evaluation of nodule detection algorithms. Any team can upload results to facilitate benchmarking. The performance of six algorithms for which results are available are compared; five from academic groups and one commercially available system. A method to combine the output of multiple systems is proposed. Results show a substantial performance difference between algorithms, and demonstrate that combining the output of algorithms leads to marked performance improvements.

Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: The CADDementia challenge

Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi-center data set. Using clinical practice as the starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimers disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with a total of 29 algorithms. The algorithms were trained on a small training set (n=30) and optionally on data from other sources (e.g., the Alzheimers Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework: http://caddementia.grand-challenge.org.

Automatic Lung Segmentation in CT Images with Accurate Handling of the Hilar Region

A fully automated and three-dimensional (3D) segmentation method for the identification of the pulmonary parenchyma in thorax X-ray computed tomography (CT) datasets is proposed. It is meant to be used as pre-processing step in the computer-assisted detection (CAD) system for malignant lung nodule detection that is being developed by the Medical Applications in a Grid Infrastructure Connection (MAGIC-5) Project. In this new approach the segmentation of the external airways (trachea and bronchi), is obtained by 3D region growing with wavefront simulation and suitable stop conditions, thus allowing an accurate handling of the hilar region, notoriously difficult to be segmented. Particular attention was also devoted to checking and solving the problem of the apparent ‘fusion’ between the lungs, caused by partial-volume effects, while 3D morphology operations ensure the accurate inclusion of all the nodules (internal, pleural, and vascular) in the segmented volume. The new algorithm was initially developed and tested on a dataset of 130 CT scans from the Italung-CT trial, and was then applied to the ANODE09-competition images (55 scans) and to the LIDC database (84 scans), giving very satisfactory results. In particular, the lung contour was adequately located in 96% of the CT scans, with incorrect segmentation of the external airways in the remaining cases. Segmentation metrics were calculated that quantitatively express the consistency between automatic and manual segmentations: the mean overlap degree of the segmentation masks is 0.96 ± 0.02, and the mean and the maximum distance between the mask borders (averaged on the whole dataset) are 0.74 ± 0.05 and 4.5 ± 1.5, respectively, which confirms that the automatic segmentations quite correctly reproduce the borders traced by the radiologist. Moreover, no tissue containing internal and pleural nodules was removed in the segmentation process, so that this method proved to be fit for the use in the framework of a CAD system. Finally, in the comparison with a two-dimensional segmentation procedure, inter-slice smoothness was calculated, showing that the masks created by the 3D algorithm are significantly smoother than those calculated by the 2D-only procedure.

Crowdsourced estimation of cognitive decline and resilience in Alzheimer's disease

Identifying accurate biomarkers of cognitive decline is essential for advancing early diagnosis and prevention therapies in Alzheimers disease. The Alzheimers disease DREAM Challenge was designed as a computational crowdsourced project to benchmark the current state‐of‐the‐art in predicting cognitive outcomes in Alzheimers disease based on high dimensional, publicly available genetic and structural imaging data. This meta‐analysis failed to identify a meaningful predictor developed from either data modality, suggesting that alternate approaches should be considered for prediction of cognitive performance.

Integrating longitudinal information in hippocampal volume measurements for the early detection of Alzheimer's disease.

BACKGROUND Structural MRI measures for monitoring Alzheimers Disease (AD) progression are becoming instrumental in the clinical practice, and more so in the context of longitudinal studies. This investigation addresses the impact of four image analysis approaches on the longitudinal performance of the hippocampal volume. METHODS We present a hippocampal segmentation algorithm and validate it on a gold-standard manual tracing database. We segmented 460 subjects from ADNI, each subject having been scanned twice at baseline, 12-month and 24month follow-up scan (1.5T, T1 MRI). We used the bilateral hippocampal volume v and its variation, measured as the annualized volume change Λ=δv/year(mm(3)/y). Four processing approaches with different complexity are compared to maximize the longitudinal information, and they are tested for cohort discrimination ability. Reference cohorts are Controls vs. Alzheimers Disease (CTRL/AD) and CTRL vs. Mild Cognitive Impairment who subsequently progressed to AD dementia (CTRL/MCI-co). We discuss the conditions on v and the added value of Λ in discriminating subjects. RESULTS The age-corrected bilateral annualized atrophy rate (%/year) were: -1.6 (0.6) for CTRL, -2.2 (1.0) for MCI-nc, -3.2 (1.2) for MCI-co and -4.0 (1.5) for AD. Combined (v, Λ) discrimination ability gave an Area under the ROC curve (auc)=0.93 for CTRL vs AD and auc=0.88 for CTRL vs MCI-co. CONCLUSIONS Longitudinal volume measurements can provide meaningful clinical insight and added value with respect to the baseline provided the analysis procedure embeds the longitudinal information.

Pleural nodule identification in low-dose and thin-slice lung computed tomography

A completely automated system for the identification of pleural nodules in low-dose and thin-slice computed tomography (CT) of the lung has been developed. The directional-gradient concentration method has been applied to the pleura surface and combined with a morphological opening-based procedure to generate a list of nodule candidates. Each nodule candidate is characterized by 12 morphological and textural features, which are analyzed by a rule-based filter and a neural classifier. This detection system has been developed and validated on a dataset of 42 annotated CT scans. The k-fold cross validation has been used to evaluate the neural classifier performance. The system performance variability due to different ground truth agreement levels is discussed. In particular, the poor 44% sensitivity obtained on the ground truth with agreement level 1 (nodules annotated by only one radiologist) with six FP per scan grows up to the 72% if the underlying ground truth is changed to the agreement level 2 (nodules annotated by two radiologists).

Hippocampal unified multi-atlas network (HUMAN): protocol and scale validation of a novel segmentation tool

In this study we present a novel fully automated Hippocampal Unified Multi-Atlas-Networks (HUMAN) algorithm for the segmentation of the hippocampus in structural magnetic resonance imaging. In multi-atlas approaches atlas selection is of crucial importance for the accuracy of the segmentation. Here we present an optimized method based on the definition of a small peri-hippocampal region to target the atlas learning with linear and non-linear embedded manifolds. All atlases were co-registered to a data driven template resulting in a computationally efficient method that requires only one test registration. The optimal atlases identified were used to train dedicated artificial neural networks whose labels were then propagated and fused to obtain the final segmentation. To quantify data heterogeneity and protocol inherent effects, HUMAN was tested on two independent data sets provided by the Alzheimers Disease Neuroimaging Initiative and the Open Access Series of Imaging Studies. HUMAN is accurate and achieves state-of-the-art performance (Dice[Formula: see text] and Dice[Formula: see text]). It is also a robust method that remains stable when applied to the whole hippocampus or to sub-regions (patches). HUMAN also compares favorably with a basic multi-atlas approach and a benchmark segmentation tool such as FreeSurfer.

MAGIC-5: an Italian mammographic database of digitised images for research

The implementation of a database of digitised mammograms is discussed. The digitised images were collected beginning in 1999 by a community of physicists in collaboration with radiologists in several Italian hospitals as a first step in developing and implementing a computer-aided detection (CAD) system. All 3,369 mammograms were collected from 967 patients and classified according to lesion type and morphology, breast tissue and pathology type. A dedicated graphical user interface was developed to visualise and process mammograms to support the medical diagnosis directly on a high-resolution screen. The database has been the starting point for developing other medical imaging applications, such as a breast CAD, currently being upgraded and optimised for use in a distributed environment with grid services, in the framework of the Instituto Nazionale di Fisicia Nucleare (INFN)-funded Medical Applications on a Grid Infrastructure Connection (MAGIC)-5 project.RiassuntoIn qesto lavoro viene discussa l’implementazione di un database immagini mammografiche digitalizzate. Le immagini sono state raccolte dal 1999 da un gruppo di fisici in collaborazione con radiology di alcuni ospedali italiani, come primo passo dello sviluppo e implementazione di un sistema di Computer Aided Detection (CAD). I 3369 mammogrammi appartengono a 967 pazienti e sono classificati secondo I tipi e la morfologia delle lesioni, il tessuto mammario e i tipi di patologie. Una interfaccia grafica opportunamente progettata è stata sviluppata per la visualizzazione e l’elaborazione delle mammografie digitalizzate al fine di runpoter supportare direttamente una diagnosi medica su monitor ad alta risoluzione. Il database ha rappresentato il punto di partenza per lo sviluppo di altre applicazioni di imaging medicale come il CAD mammografico costantemente ottimizzato e aggiornato con l’uso di un ambiente distribuito che dispone di servizi GRID, nel framework del progetto MAGIC-5, finanziato dell’INFN.

GPCALMA: an italian mammographic database of digitized images for research

In this work the implementation of a database of digitized mammograms is described. The digitized images were collected since 1999 by a community of physicists in collaboration with radiologists in several Italian hospitals, as a first step in order to develop and implement a Computer Aided Detection (CAD) system. 3369 mammograms were collected from 967 patients; they were classified according to the type and the morphology of the lesions, the type of the breast tissue and the type of pathologies. A dedicated Graphical User Interface was developed for mammography visualization and processing, in order to support the medical diagnosis directly on a high-resolution screen. The database has been the starting point for the development of other medical imaging applications such as a breast CAD, currently being upgraded and optimized for the use in conjunction of the GRID technology in the framework of the INFN-funded MAGIC-5 project.

Diagnostic performance of radiologists with and without different CAD systems for mammography

The purpose of this study is the evaluation of the variation of performance in terms of sensitivity and specificity of two radiologists with different experience in mammography, with and without the assistance of two different CAD systems. The CAD considered are SecondLookTM (CADx Medical Systems, Canada), and CALMA (Computer Assisted Library in MAmmography). The first is a commercial system, the other is the result of a research project, supported by INFN (Istituto Nazionale di Fisica Nucleare, Italy); their characteristics have already been reported in literature. To compare the results with and without these tools, a dataset composed by 70 images of patients with cancer (biopsy proven) and 120 images of healthy breasts (with a three years follow up) has been collected. All the images have been digitized and analysed by two CAD, then two radiologists with respectively 6 and 2 years of experience in mammography indipendently made their diagnosis without and with, the support of the two CAD systems. In this work sensitivity and specificity variation, the Az area under the ROC curve, are reported. The results show that the use of a CAD allows for a substantial increment in sensitivity and a less pronounced decrement in specificity. The extent of these effects depends on the experience of the readers and is comparable for the two CAD considered.