BIOIMAGING 2017 Abstracts


Full Papers
Paper Nr: 6
Title:

Segmentation of Retinal Ganglion Cells From Fluorescent Microscopy Imaging

Authors:

Silvia Baglietto, Ibolya E. Kepiro, Gerrit Hilgen, Evelyne Sernagor, Vittorio Murino and Diego Sona

Abstract: The visual information processing starts in the retina. The working mechanisms of its complex stratified circuits, in which ganglion cells play a central role, is still largely unknown. Understanding the visual coding is a challenging and active research area also requiring automated analysis of retinal images. It demands appropriate algorithms and methods for studying a network population of strictly entangled cells. Within this framework, we propose a combined technique for segmenting retinal ganglion cell (RGC) bodies, the output elements of the retina. The method incorporates a blob enhancement filtering in order to select the specific cell shapes, an active contour process for precise border segmentation and a watershed transform step which separates single cell contours in possible grouped segmentations. The proposed approach has been validated on fluorescent microscopy images of mouse retinas with promising results.

Paper Nr: 8
Title:

Thrombus Detection in CT Brain Scans using a Convolutional Neural Network

Authors:

Aneta Lisowska, Erin Beveridge, Keith Muir and Ian Poole

Abstract: Automatic detection and measurement of thrombi may expedite clinical workflow in the treatment planning stage. Nevertheless it is a challenging task on non-contrast computed tomography due to the subtlety of the pathological intensity changes, which are further confounded by the appearance of vascular calcification (common in ageing brains). In this paper we propose a 3D Convolutional Neural Network architecture to detect these subtle signs of stroke. The architecture is designed to exploit contralateral features and anatomical atlas information. We use 122 CT volumes equally split into training and testing to validate our method, achieving a ROC AUC of 0.996 and a Precision-Recall AUC of 0.563 in a voxel-level evaluation. The results are not yet at a level for routine clinical use, but they are encouraging.

Paper Nr: 15
Title:

Real-time Display and In-vivo Optical-resolution Photoacoustic Microscopy for Ophthalmic Imaging

Authors:

Sang-Won Lee, Heesung Kang and Tea Geol Lee

Abstract: Photoacoustic imaging is a non-invasive imaging technology that can be combined with optical absorption contrast and detection of acoustic wave for structural, functional, and molecular imaging. Especially, optical-resolution photoacoustic microscopy (OR-PAM) can provide a high spatial resolution with a micron-scale. In this study, we have developed laser-scanning OR-PAM, which could obtain in-vivo photoacoustic ophthalmic angiography. For high speed image acquisition, we used a nanosecond pulsed laser with a 300 kHz-pulse repetition rates. In addition, we carried out parallel signal processing using a graphics processing unit to enable fast signal processing. Therefore, we successfully obtained maximum amplitude projection images of microvasculature in anterior and posterior segments of mouse’s eye with real-time display of 0.98 fps.

Paper Nr: 22
Title:

Real-time Anterior Mitral Leaflet Tracking using Morphological Operators and Active Contours

Authors:

Malik Saad Sultan, Nelson Martins, Eva Costa, Diana Veiga, Manuel João Ferreira, Sandra Mattos and Miguel Tavares Coimbra

Abstract: The mitral valve plays a vital role in our circulatory system. To study its functionality, it is important to measure clinically relevant parameters, such as its thickness, mobility and shape. Since manual segmentation is impractical, time consuming and requires expert knowledge, an automatic segmentation tool can have a significant clinical impact, providing objective measures to clinicians for understanding the morphology and behaviour of the mitral valve. In this work, a real time tracking method has been proposed for ultrasound videos obtained with the Parasternal Long Axis view. The algorithm is semi-automatic, assumes manual Anterior Mitral Leaflet segmentation in the first frame and then it uses mathematical morphology algorithms to obtain tracking results, further refined by localized active contours during the whole cardiac cycle. Finally, the medial axis is extracted for a quantitative analysis. Results show that the algorithm can segment 1137 frames extracted from 9 fully annotated sequences of the real clinical video data in only 0.89 sec/frame, with an average error of 5 pixels. Furthermore, the algorithms exhibited robust tracking performance in the most difficult situations, which are large frame-to-frame displacements.

Paper Nr: 24
Title:

A Preliminary Approach for the Segmentation of Mitral Valve Regurgitation Jet in Doppler Ecocardiography Images

Authors:

Eva Costa, Nelson Martins, Malik Saad Sultan, Diana Veiga, Manuel João Ferreira, Sandra Mattos and Miguel Coimbra

Abstract: Rheumatic Fever and Rheumatic Heart Disease remain a major burden among children in developing countries. Echocardiography with colour flow Doppler is key to early diagnosis. However, the technique requires time and experienced operators, which are scarce resources in the affected areas. Automatic segmentation of colour Doppler regurgitation jets could, potentially, reduce the cost of screening, and spread diagnostic accessibility for a larger number of patients. Ultrasound processing is very challenging due to speckle noise and similarity of representation of all kinds of tissue. Region-based active contours are suitable tools for the segmentation in cases of intensity heterogeneities, which makes them interesting algorithms for left atrium segmentation. HSV colour space describes colour in terms of hues and saturation, which may facilitate the translation of medical interpretation of the Doppler pseudo-colour into mathematical expression for colour segmentation. A total of 979 frames from 20 sequences were manually annotated and used to validate the proposed pipeline. Overall, the results for colour pattern segmentation are promising (sensitivity=0.91 false detection rate=0.10), but further developments are required for the atrium segmentation (sensitivity=0.80, false detection rate=0.28).

Paper Nr: 29
Title:

Localization of Demyelinating Plaques in MRI using Convolutional Neural Networks

Authors:

Bartłomiej Stasiak, Pawel Tarasiuk, Izabela Michalska, Arkadiusz Tomczyk and Piotr S. Szczepaniak

Abstract: In the paper a method of demyelinating plaques localization in head MRI sequences is presented. For that purpose a convolutional neural network is used. It is trained to act as non-linear filter, which should indicate (give a high response) in those image areas where the sought objects are located. Consequently, the output of the proposed architecture is an image and not a single label as it is in the case of traditional networks with pooling and fully connected layers. Another interesting feature of the proposed solution is the ability to select network parameters using smaller patches cut from training images which reduces the amount of data that must be propagated through the network. It should be emphasized that the conducted research was possible only thanks to the manually outlined plaques provided by radiologist.

Short Papers
Paper Nr: 5
Title:

fMRI and Voxel-based Morphometry in Detection of Early Stages of Alzheimer's Disease

Authors:

Andrey V. Sokolov, Sergey V. Vorobyev, Aleksandr Yu. Efimtcev, Viacheslav S. Dekan, Gennadiy E. Trufanov, Vladimir Yu. Lobzin and Vladimir A. Fokin

Abstract: Alzheimer’s disease (AD) is the most common form of dementia in older adults. Loss of memory is the usual first symptom and different brain regions are involved to this pathological process. The aim of the study was to investigate the organization of cortical areas responsible for visual memory and determine correlation between memory impairment and atrophy of memory specific brain regions in early stages of AD. Voxel-based MR-morphometry was used to evaluate brain atrophy and functional MRI was used to detect specific brain regions responsible to visual memory task in patients with Alzheimer's disease and in control group. FMRI was performed on Siemens Magnetom Symphony (1.5 T ) with the use of Blood Oxygenation Level Dependent technique (BOLD), based on distinctions of magnetic properties of hemoglobin. For test stimuli we used blocks of 12 not related images for "Baseline" and 12 images with 6 presented before for "Active". Stimuli were presented 3 times with reduction of repeated images to 4 and 2. For functional and morthometric data post-processing we used SPM8. Patients with Alzheimer's disease showed less activation in hippocampal formation (HF) region and parahippocampal gyrus then the control group (p<0.05). The study also showed reduced activation in posterior cingulate cortex (p<0.001). Voxel-based morphometry showed significant atrophy of grey matter in Alzheimer’s disease patients, especially of both temporal lobes (fusiform and parahippocampal gyri); frontal lobes (posterior cingulate and superior frontal gyri). The study showed correlation between memory impairment and atrophy of memory specific brain regions of frontal and medial temporal lobes. Reduced activation in hippocampal formation and parahippocampal gyri, in posterior cingulate gyrus in patients with Alzheimer's disease correlates to significant atrophy of these regions, detected by voxel-based morphometry. The use of functional MRI and voxel-based morphometry provides the way to find alterations in brain function on early stages of AD before the development of significant irreversible structural damage.

Paper Nr: 7
Title:

Automatic Classification of Z-ring Formation Stages at the Single Cell Level in Escherichia Coli by Machine Learning

Authors:

Marzieh Zare, Ramakanth Neeli-Venkata, Leonardo Martins, Sari Peltonen, Ulla Ruotsalainen and Andre S. Ribeiro

Abstract: In E. coli, Z-ring formation precedes the assembly of the membrane that partitions a cell into two daughter cells. Initially, as FtsZ proteins are expressed, they preferentially locate at the poles. After, they form a ring at midcell, in between the nucleoids, ‘marking’ where a constriction will form. Finally, the ring becomes a circle, where the septum separating the daughter cells forms. Being the temporal-spatial organization of FtsZ noisy, differing between cells in timing and location, its study requires observing many cells by time-lapse microscopy. To assist, image and signal processing methods are needed to extract information unbiasedly from many cells. Also, one needs automatic identification of the ring formation stage in individual cells. Here we used three classification methods to identify the stage of ring formation from microscopy images: Decision Tree (DT), Support Vector Machine (SVM), and Regularized Multinomial Logistic regression (RMLR). We find that RMLR performs better (higher 10-fold cross-validated accuracy, ACC). Our study will assist future studies at the single cell level of the spatio-temporal dynamics of cell division in E. coli.

Paper Nr: 9
Title:

Automated Segmentation of Upper Airways from MRI - Vocal Tract Geometry Extraction

Authors:

Antti Ojalammi and Jarmo Malinen

Abstract: An algorithm for automatically extracting a triangulated surface mesh of the human vocal tract from 3D MRI data is proposed. The algorithm is based on a combination of anatomic landmarking, seeded region growing, and smoothing. Using these methods, a mask is automatically created for removing unwanted details not associated with the vocal tract from the MRI voxel data. The mask is then applied to the original MRI data, after which marching cubes algorithm is used for extracting a triangulated surface. The proposed method can be used for processing large datasets, e.g., for validation of numerical methods in speech sciences as well as for anatomical studies.

Paper Nr: 13
Title:

Boosted Tree Classifier for in Vivo Identification of Early Cervical Cancer using Multispectral Digital Colposcopy

Authors:

Nilgoon Zarei, Dennis Cox, Pierre Lane, Scott Cantor, Neely Atkinson, Jose-Miguel Yamal, Leonid Fradkin, Daniel Serachitopol, Sylvia Lam, Dirk Niekerk, Dianne Miller, Jessica McAlpine, Kayla Castaneda, Felipe Castaneda, Michele Follen and Calum MacAulay

Abstract: Background: Cervical cancer develops over several years; screening and early diagnosis have decreased the incidence and mortality threefold over the last fifty years. Opportunities for the application of imaging and automation in the screening process exist in settings where resources are limited. Methods: Patients with high-grade squamous intraepithelial lesions (SIL) underwent imaging with a Multispectral Digital Colposcopy (MDC) prior to have a loop excision of the cervix. The image taken with white light was annotated by a clinician. The excised specimen was mapped by the study histopathologist blinded to the MDC data. This map was used to define areas of high grade in the excised tissue. Eleven reviewers mapped the histopathologic data into the MDC images. The reviewers’ maps were analyzed and areas of agreement were calculated. We compared the result of a boosted tree classifier with a previously developed ensemble classifier. Results: Using a boosted tree classifier we obtained a sensitivity of 95%, a specificity of 96%, and an accuracy of 96% on the training sets. When we applied the classifier to a test set, we obtained a sensitivity of 82%, a specificity of 81%, and an accuracy of 81%. The boosted tree classifier performed better than the previously developed ensemble classifier. Conclusion: Here we presented promising results which show that a boosted tree analysis on MDC images is a method that could be used as an adjunct to colposcopy and would result in greater diagnostic accuracy compared to existing methods.

Paper Nr: 19
Title:

Automatic Quantification of Vocal Cord Paralysis - An Application of Fibre-optic Endoscopy Video Processing

Authors:

Radhika Menon, Lykourgos Petropoulakis, John J. Soraghan, Heba Lakany, Kenneth MacKenzie, Omar Hilmi and Gaetano Di Caterina

Abstract: Full movement of the vocal cords is necessary for life sustaining functions. To enable correct diagnosis of reduced vocal cord motion and thereby potentially enhance treatment outcomes, it is proposed to objectively determine the degree of vocal cord paralysis in contrast to the current clinical practice of subjective evaluation. Our study shows that quantitative assessment can be achieved using optical flow based motion estimation of the opening and closing movements of the vocal cords. The novelty of the proposed method lies in the automatic processing of fibre-optic endoscopy videos to derive an objective measure for the degree of paralysis, without the need for high-end data acquisition systems such as high speed cameras or stroboscopy. Initial studies with three video samples yield promising results and encourage further investigation of vocal cord paralysis using this technique.

Paper Nr: 20
Title:

A Multi-modal Brain Image Registration Framework for US-guided Neuronavigation Systems - Integrating MR and US for Minimally Invasive Neuroimaging

Authors:

Francesco Ponzio, Enrico Macii, Elisa Ficarra and Santa Di Cataldo

Abstract: US-guided neuronavigation exploits the simplicity of use and minimal invasiveness of Ultrasound (US) imaging and the high tissue resolution and signal-to-noise ratio of Magnetic Resonance Imaging (MRI) to guide brain surgeries. More specifically, the intra-operative 3D US images are combined with pre-operative MR images to accurately localise the course of instruments in the operative field with minimal invasiveness. Multimodal image registration of 3D US and MR images is an essential part of such system. In this paper, we present a complete software framework that enables the registration US and MR brain scans based on a multi resolution deformable transform, tackling elastic deformations (i.e. brain shifts) possibly occurring during the surgical procedure. The framework supports also simpler and faster registration techniques, based on rigid or affine transforms, and enables the interactive visualisation and rendering of the overlaid US and MRI volumes. The registration was experimentally validated on a public dataset of realistic brain phantom images, at different levels of artificially induced deformations.

Paper Nr: 25
Title:

Evaluating Spatial Coverage of Breast Examination with Free-hand Ultrasound Transducer

Authors:

Zuzana Bílková, Michal Bartoš, Jan Schier, Filip Šroubek, Barbara Zitová, Jan Vydra and Jan Daneš

Abstract: Ultrasound examination plays an important role in both breast cancer screening and diagnostics. One of the drawbacks of the US examination is the uncertainty whether the whole breast was scanned. The proposed paper addresses the methodology how the completeness of the examination can be efficiently evaluated. We propose an affordable solution for simultaneously tracking and grabbing a video from a free-hand 2D ultrasound transducer during standard breast examinations by means of the probe motion tracking. From the recorded data we calculate duration in seconds, for which every part of the examined region has been captured and perform algorithmically local 3D reconstruction. Thus the system can inform the specialist performing the exam about regions that were insufficiently examined and minimize the risk of not detecting developing harmful lesions. The measure for the evaluation and comparison of the individual examinations is proposed. The functionality of the method is illustrated.

Paper Nr: 26
Title:

Brain Tumor Segmentation Based on Non Negative Matrix Factorization and Fuzzy Clustering

Authors:

Ujjwal Baid, Shubham Talbar and Sanjay Talbar

Abstract: The problem of computational brain tumor segmentation has attracted researchers over a decade because of its high clinical relevance and challenging nature. Automatic and accurate detection of brain tumor is one of the major areas of research in medical image processing which helps radiologists for precise treatment planning. Magnetic Resonance Imaging (MRI) is one of the widely used imaging modality for visualizing and assessing the brain anatomy and its functions in non-invasive manner. In this paper a novel approach for brain tumor segmentation based on Non-Negative Matrix Factorization(NMF) and Fuzzy clustering is proposed. Proposed algorithm is tested on BRATS 2012 training database of High Grade and Low Grade Glioma tumors with clinical and synthetic data of 80 patients. Various evaluation parameters like Dice index, Jaccard index, Sensitivity, Specificity are evaluated. Comparison of experimental results with other state of the art brain tumor segmentation methods demonstrate that proposed method outperforms existing segmentation techniques.

Posters
Paper Nr: 14
Title:

Automated Cell Segmentation of Fission Yeast Phase Images - Segmenting Cells from Light Microscopy Images

Authors:

Jennifer O'Brien, Sanaul Hoque, Daniel Mulvihill and Konstantinos Sirlantzis

Abstract: Robust image analysis is an important aspect of all cell biology studies. The geometrics of cells are critical for developing an understanding of biological processes. Time constraints placed on researchers lead to a narrower focus on what data are collected and recorded from an experiment, resulting in a loss of data. Currently, preprocessing of microscope images is followed by the utilisation and parameterisation of inbuilt functions of various softwares to obtain information. Using the fission yeast, Schizosaccharomyes pombe, we propose a novel, fully automated, segmentation software for cells with a significantly lower rate of segmentation errors than PombeX with the same dataset.

Paper Nr: 18
Title:

A Non-Local Diffusion Saliency Model for Magnetic Resonance Imaging

Authors:

I. Ramírez, G. Galiano, N. Malpica and E. Schiavi

Abstract: Based on previous work on image classification and recent applications of non-local non-linear diffusion equations, we propose a non-local p-laplacian variational model for saliency detection in digital images. Focusing on the range 0 < p < 1 we also consider the regularized non-convex fluxes generated by the related hyper-laplacian diffusion operators. With the aim of exploring the properties and potential applications of such non-local, non-convex operators the model is applied to Magnetic Resonace Imaging (MRI) for Fluid Attenuated Inversion Recovery image (FLAIR) modality showing promising numerical results. In this work Saliency shall be understood as the relevant, outstanding region in a FLAIR image, which is commonly the brightest part. It corresponds to a tumor and neighborhood edema. Our preliminary experiments show that the proposed model can achieve very accurate results in this modality in terms of all the considered metrics.

Paper Nr: 23
Title:

Understanding the Energy Saving Potential of Smart Scale Selection in the Viola and Jones Facial Detection Algorithm

Authors:

Noel Perez, Sérgio Faria and Miguel Coimbra

Abstract: In this paper we study the energy saving potential of smart scale selection methods when using the Viola and Jones face detector running on smartphone devices. Our motivation is that cloud and edge-cloud multi-user environments may provide enough contextual information to create this type of scale selection algorithms. Given their non-trivial design, we must first inspect its actual benefits, before committing important research resources to actually produce relevant smart scale selection methods. Our experimental methodology in this paper assumes the optimum scenario of a perfect selection of scales for each image (drawn from ground truth annotation, using well-known public datasets), comparing it with the typical multi-scale geometrical progression approach of the Viola Jones algorithm, measuring both classification precision and recall, as well as algorithmic execution time and battery consumption on Android smartphone devices. Results show that if we manage to approximate this perfect scale selection, we obtain very significant energy savings, motivating a strong research investment on this topic.

Paper Nr: 27
Title:

Saliency Guided Computer-aided Diagnosis for Neurodegenerative Dementia

Authors:

Olfa Ben Ahmed, Mohamed-Chacker Larabi, Marc Paccalin and Christine Fernandez-Maloigne

Abstract: Visual assessment of brain atrophy for brain diseases diagnosis by clinicians is the most widely adopted method in clinical practices. Such a visually extracted knowledge represents a great potential to develop better training programs and create new tools to assist clinical decision making. Inspired by the clinician visual behavior, we propose in this work a new and automatic approach to detect and quantify local brain atrophies. The proposed approach combines both bottom-up and top-down visual saliency using domain knowledge in the brain MRI analysis. The first subsystem relies on low-level MRI characterization (texture and edge) while the second is based on an embedded learning process to identify and localize the subset of gray matter regions that provides optimal discrimination between subjects. The proposed method validated for the task of Alzheimer’s disease (AD) subjects recognition. Classification experiments were conducted on a subset of 188 anatomical MR images extracted from the Alzheimer’s Disease Neuro-imaging Initiative (ADNI) dataset. We report accuracy of 81.48% and 76.66% respectively for AD versus Normal Control (NC) and Mild Cognitive Impairment (MCI) versus NC classification tasks.