Daniel Jimenez-Carretero

A Benchmark for Comparison of Cell Tracking Algorithms

Motivation: Automatic tracking of cells in multidimensional time-lapse fluorescence microscopy is an important task in many biomedical applications. A novel framework for objective evaluation of cell tracking algorithms has been established under the auspices of the IEEE International Symposium on Biomedical Imaging 2013 Cell Tracking Challenge. In this article, we present the logistics, datasets, methods and results of the challenge and lay down the principles for future uses of this benchmark. Results: The main contributions of the challenge include the creation of a comprehensive video dataset repository and the definition of objective measures for comparison and ranking of the algorithms. With this benchmark, six algorithms covering a variety of segmentation and tracking paradigms have been compared and ranked based on their performance on both synthetic and real datasets. Given the diversity of the datasets, we do not declare a single winner of the challenge. Instead, we present and discuss the results for each individual dataset separately. Availability and implementation: The challenge Web site (http://www.codesolorzano.com/celltrackingchallenge) provides access to the training and competition datasets, along with the ground truth of the training videos. It also provides access to Windows and Linux executable files of the evaluation software and most of the algorithms that competed in the challenge. Contact: codesolorzano@unav.es Supplementary information: Supplementary data are available at Bioinformatics online.

Comparing algorithms for automated vessel segmentation in computed tomography scans of the lung: the VESSEL12 study

The VESSEL12 (VESsel SEgmentation in the Lung) challenge objectively compares the performance of different algorithms to identify vessels in thoracic computed tomography (CT) scans. Vessel segmentation is fundamental in computer aided processing of data generated by 3D imaging modalities. As manual vessel segmentation is prohibitively time consuming, any real world application requires some form of automation. Several approaches exist for automated vessel segmentation, but judging their relative merits is difficult due to a lack of standardized evaluation. We present an annotated reference dataset containing 20 CT scans and propose nine categories to perform a comprehensive evaluation of vessel segmentation algorithms from both academia and industry. Twenty algorithms participated in the VESSEL12 challenge, held at International Symposium on Biomedical Imaging (ISBI) 2012. All results have been published at the VESSEL12 website http://vessel12.grand-challenge.org. The challenge remains ongoing and open to new participants. Our three contributions are: (1) an annotated reference dataset available online for evaluation of new algorithms; (2) a quantitative scoring system for objective comparison of algorithms; and (3) performance analysis of the strengths and weaknesses of the various vessel segmentation methods in the presence of various lung diseases.

3D Frangi-based lung vessel enhancement filter penalizing airways

This paper describes a fully automatic simultaneous lung vessel and airway enhancement filter. The approach consists of a Frangi-based multiscale vessel enhancement filtering specifically designed for lung vessel and airway detection, where arteries and veins have high contrast with respect to the lung parenchyma, and airway walls are hollow tubular structures with a non negative response using the classical Frangis filter. The features extracted from the Hessian matrix are used to detect centerlines and approximate walls of airways, decreasing the filter response in those areas by applying a penalty function to the vesselness measure. We validate the segmentation method in 20 CT scans with different pathological states within the VESSEL12 challenge framework. Results indicate that our approach obtains good results, decreasing the number of false positives in airway walls.

Optimal multiresolution 3D level-set method for liver segmentation incorporating local curvature constraints

Advanced liver surgery requires a precise pre-operative planning, where liver segmentation and remnant liver volume are key elements to avoid post-operative liver failure. In that context, level-set algorithms have achieved better results than others, especially with altered liver parenchyma or in cases with previous surgery. In order to improve functional liver parenchyma volume measurements, in this work we propose two strategies to enhance previous level-set algorithms: an optimal multi-resolution strategy with fine details correction and adaptive curvature, as well as an additional semiautomatic step imposing local curvature constraints. Results show more accurate segmentations, especially in elongated structures, detecting internal lesions and avoiding leakages to close structures.

Organ-focused mutual information for nonrigid multimodal registration of liver CT and Gd–EOB–DTPA-enhanced MRI

Accurate detection of liver lesions is of great importance in hepatic surgery planning. Recent studies have shown that the detection rate of liver lesions is significantly higher in gadoxetic acid-enhanced magnetic resonance imaging (Gd-EOB-DTPA-enhanced MRI) than in contrast-enhanced portal-phase computed tomography (CT); however, the latter remains essential because of its high specificity, good performance in estimating liver volumes and better vessel visibility. To characterize liver lesions using both the above image modalities, we propose a multimodal nonrigid registration framework using organ-focused mutual information (OF-MI). This proposal tries to improve mutual information (MI) based registration by adding spatial information, benefiting from the availability of expert liver segmentation in clinical protocols. The incorporation of an additional information channel containing liver segmentation information was studied. A dataset of real clinical images and simulated images was used in the validation process. A Gd-EOB-DTPA-enhanced MRI simulation framework is presented. To evaluate results, warping index errors were calculated for the simulated data, and landmark-based and surface-based errors were calculated for the real data. An improvement of the registration accuracy for OF-MI as compared with MI was found for both simulated and real datasets. Statistical significance of the difference was tested and confirmed in the simulated dataset (p<0.01).

Automated axial right ventricle to left ventricle diameter ratio computation in computed tomography pulmonary angiography.

Background and Purpose Right Ventricular to Left Ventricular (RV/LV) diameter ratio has been shown to be a prognostic biomarker for patients suffering from acute Pulmonary Embolism (PE). While Computed Tomography Pulmonary Angiography (CTPA) images used to confirm a clinical suspicion of PE do include information of the heart, a numerical RV/LV diameter ratio is not universally reported, likely because of lack in training, inter-reader variability in the measurements, and additional effort by the radiologist. This study designs and validates a completely automated Computer Aided Detection (CAD) system to compute the axial RV/LV diameter ratio from CTPA images so that the RV/LV diameter ratio can be a more objective metric that is consistently reported in patients for whom CTPA diagnoses PE. Materials and Methods The CAD system was designed specifically for RV/LV measurements. The system was tested in 198 consecutive CTPA patients with acute PE. Its accuracy was evaluated using reference standard RV/LV radiologist measurements and its prognostic value was established for 30-day PE-specific mortality and a composite outcome of 30-day PE-specific mortality or the need for intensive therapies. The study was Institutional Review Board (IRB) approved and HIPAA compliant. Results The CAD system analyzed correctly 92.4% (183/198) of CTPA studies. The mean difference between automated and manually computed axial RV/LV ratios was 0.03±0.22. The correlation between the RV/LV diameter ratio obtained by the CAD system and that obtained by the radiologist was high (r=0.81). Compared to the radiologist, the CAD system equally achieved high accuracy for the composite outcome, with areas under the receiver operating characteristic curves of 0.75 vs. 0.78. Similar results were found for 30-days PE-specific mortality, with areas under the curve of 0.72 vs. 0.75. Conclusions An automated CAD system for determining the CT derived RV/LV diameter ratio in patients with acute PE has high accuracy when compared to manual measurements and similar prognostic significance for two clinical outcomes.

Automatic ventricle detection in Computed Tomography Pulmonary Angiography

Automated medical image analysis requires methods to localize anatomic structures in the presence of normal interpatient variability, pathology, and the different protocols used to acquire images for different clinical settings. Recent advances have improved object detection in the context of natural images, but they have not been adapted to the 3D context of medical images. In this paper we present a 2.5D object detector designed to locate, without any user interaction, the left and right heart ventricles in Computed Tomography Pulmonary Angiography (CTPA) images. A 2D object detector is trained to find ventricles on axial slices. Those detections are automatically clustered according to their size and position. The cluster with highest score, representing the 3D location of the ventricle, is then selected. The proposed method is validated in 403 CTPA studies obtained in patients with clinically suspected pulmonary embolism. Both ventricles are properly detected in 94.7% of the cases. The proposed method is very generic and can be easily adapted to detect other structures in medical images.

Liver Segmentation and Volume Estimation from Preoperative CT Images in Hepatic Surgical Planning: Application of a Semiautomatic Method Based on 3D Level Sets

Laura Fernandez-de-Manuel1,5, Maria J. Ledesma-Carbayo1,5, Daniel Jimenez-Carretero1,5, Javier Pascau2, Jose L. Rubio-Guivernau1,5, Jose M. Tellado3, Enrique Ramon4, Manuel Desco2 and Andres Santos1,5 1Biomedical Image Technologies Lab, Universidad Politecnica de Madrid 2Medicina y Cirugia Experimental, Hospital General Universitario Gregorio Maranon 3 Servicio de Cirugia General I, Hospital General Universitario Gregorio Maranon 4 Servicio de Radiodiagnostico, Hospital General Universitario Gregorio Maranon, Madrid 5Biomedical Research Center in Bioengineering, Biomaterials, and Nanomedicine (CIBER-BBN), Madrid Spain

ESC-Track: A computer workflow for 4-D segmentation, tracking, lineage tracing and dynamic context analysis of ESCs

Embryonic stem cells (ESCs) can be established as permanent cell lines, and their potential to differentiate into adult tissues has led to widespread use for studying the mechanisms and dynamics of stem cell differentiation and exploring strategies for tissue repair. Imaging live ESCs during development is now feasible due to advances in optical imaging and engineering of genetically encoded fluorescent reporters; however, a major limitation is the low spatio-temporal resolution of long-term 3-D imaging required for generational and neighboring reconstructions. Here, we present the ESC-Track (ESC-T) workflow, which includes an automated cell and nuclear segmentation and tracking tool for 4-D (3-D + time) confocal image data sets as well as a manual editing tool for visual inspection and error correction. ESC-T automatically identifies cell divisions and membrane contacts for lineage tree and neighborhood reconstruction and computes quantitative features from individual cell entities, enabling analysis of fluorescence signal dynamics and tracking of cell morphology and motion. We use ESC-T to examine Myc intensity fluctuations in the context of mouse ESC (mESC) lineage and neighborhood relationships. ESC-T is a powerful tool for evaluation of the genealogical and microenvironmental cues that maintain ESC fitness.

Deep-learning strategy for pulmonary artery-vein classification of non-contrast CT images

Artery-vein classification on pulmonary computed tomography (CT) images is becoming of high interest in the scientific community due to the prevalence of pulmonary vascular disease that affects arteries and veins through different mechanisms. In this work, we present a novel approach to automatically segment and classify vessels from chest CT images. We use a scale-space particle segmentation to isolate vessels, and combine a convolutional neural network (CNN) to graph-cut (GC) to classify the single particles. Information about proximity of arteries to airways is learned by the network by means of a bronchus enhanced image. The methodology is evaluated on the superior and inferior lobes of the right lung of twenty clinical cases. Comparison with manual classification and a Random Forests (RF) classifier is performed. The algorithm achieves an overall accuracy of 87% when compared to manual reference, which is higher than the 73% accuracy achieved by RF.