A Dataset of Annotated Omnidirectional Videos for Distancing Applications.

Omnidirectional (or 360°) cameras are acquisition devices that, in the next few years, could have a big impact on video surveillance applications, research, and industry, as they can record a spherical view of a whole environment from every perspective. This paper presents two new contributions to the research community: the CVIP360 dataset, an annotated dataset of 360° videos for distancing applications, and a new method to estimate the distances of objects in a scene from a single 360° image. The CVIP360 dataset includes 16 videos acquired outdoors and indoors, annotated by adding information about the pedestrians in the scene (bounding boxes) and the distances to the camera of some points in the 3D world by using markers at fixed and known intervals. The proposed distance estimation algorithm is based on geometry facts regarding the acquisition process of the omnidirectional device, and is uncalibrated in practice: the only required parameter is the camera height. The proposed algorithm was tested on the CVIP360 dataset, and empirical results demonstrate that the estimation error is negligible for distancing applications.

A Novel Solution Based on Scale Invariant Feature Transform Descriptors and Deep Learning for the Detection of Suspicious Regions in Mammogram Images.

BACKGROUND: Deep learning methods have become popular for their high-performance rate in the classification and detection of events in computer vision tasks. Transfer learning paradigm is widely adopted to apply pretrained convolutional neural network (CNN) on medical domains overcoming the problem of the scarcity of public datasets. Some investigations to assess transfer learning knowledge inference abilities in the context of mammogram screening and possible combinations with unsupervised techniques are in progress. METHODS: We propose a novel technique for the detection of suspicious regions in mammograms that consist of the combination of two approaches based on scale invariant feature transform (SIFT) keypoints and transfer learning with pretrained CNNs such as PyramidNet and AlexNet fine-tuned on digital mammograms generated by different mammography devices. Preprocessing, feature extraction, and selection steps characterize the SIFT-based method, while the deep learning network validates the candidate suspicious regions detected by the SIFT method. RESULTS: The experiments conducted on both mini-MIAS dataset and our new public dataset Suspicious Region Detection on Mammogram from PP (SuReMaPP) of 384 digital mammograms exhibit high performances compared to several state-of-the-art methods. Our solution reaches 98% of sensitivity and 90% of specificity on SuReMaPP and 94% of sensitivity and 91% of specificity on mini-MIAS. CONCLUSIONS: The experimental sessions conducted so far prompt us to further investigate the powerfulness of transfer learning over different CNNs and possible combinations with unsupervised techniques. Transfer learning performances' accuracy may decrease when the training and testing images come out from mammography devices with different properties.

Convolutional architectures for virtual screening.

BACKGROUND: A Virtual Screening algorithm has to adapt to the different stages of this process. Early screening needs to ensure that all bioactive compounds are ranked in the first positions despite of the number of false positives, while a second screening round is aimed at increasing the prediction accuracy. RESULTS: A novel CNN architecture is presented to this aim, which predicts bioactivity of candidate compounds on CDK1 using a combination of molecular fingerprints as their vector representation, and has been trained suitably to achieve good results as regards both enrichment factor and accuracy in different screening modes (98.55% accuracy in active-only selection, and 98.88% in high precision discrimination). CONCLUSION: The proposed architecture outperforms state-of-the-art ML approaches, and some interesting insights on molecular fingerprints are devised.

Image Content Enhancement Through Salient Regions Segmentation for People With Color Vision Deficiencies.

Color vision deficiencies affect visual perception of colors and, more generally, color images. Several sciences such as genetics, biology, medicine, and computer vision are involved in studying and analyzing vision deficiencies. As we know from visual saliency findings, human visual system tends to fix some specific points and regions of the image in the first seconds of observation summing up the most important and meaningful parts of the scene. In this article, we provide some studies about human visual system behavior differences between normal and color vision-deficient visual systems. We eye-tracked the human fixations in first 3 seconds of observation of color images to build real fixation point maps. One of our contributions is to detect the main differences between the aforementioned human visual systems related to color vision deficiencies by analyzing real fixation maps among people with and without color vision deficiencies. Another contribution is to provide a method to enhance color regions of the image by using a detailed color mapping of the segmented salient regions of the given image. The segmentation is performed by using the difference between the original input image and the corresponding color blind altered image. A second eye-tracking of color blind people with the images enhanced by using recoloring of segmented salient regions reveals that the real fixation points are then more coherent (up to 10%) with the normal visual system. The eye-tracking data collected during our experiments are in a publicly available dataset called Eye-Tracking of Color Vision Deficiencies.

An enhanced random walk algorithm for delineation of head and neck cancers in PET studies.

An algorithm for delineating complex head and neck cancers in positron emission tomography (PET) images is presented in this article. An enhanced random walk (RW) algorithm with automatic seed detection is proposed and used to make the segmentation process feasible in the event of inhomogeneous lesions with bifurcations. In addition, an adaptive probability threshold and a k-means based clustering technique have been integrated in the proposed enhanced RW algorithm. The new threshold is capable of following the intensity changes between adjacent slices along the whole cancer volume, leading to an operator-independent algorithm. Validation experiments were first conducted on phantom studies: High Dice similarity coefficients, high true positive volume fractions, and low Hausdorff distance confirm the accuracy of the proposed method. Subsequently, forty head and neck lesions were segmented in order to evaluate the clinical feasibility of the proposed approach against the most common segmentation algorithms. Experimental results show that the proposed algorithm is more accurate and robust than the most common algorithms in the literature. Finally, the proposed method also shows real-time performance, addressing the physician's requirements in a radiotherapy environment.

A tunable digital ishihara plate for pre-school aged children.

Colors play a fundamental role for children, both in the everyday life and in education. They recognize the surrounding world, and play games making a large use of colors. They learn letters and numbers by means of colors. As a consequence, early diagnosis of color blindness is an crucial to support an individual affected by this visual perception alteration at the initial phase of his/her life. The diagnosis of red-green color deficiencies (protanopia or deuteranopia) is commonly accomplished by means of the Ishihara test, which consists of plates showing dots with different sizes where some of them compose numbers within a gamut of colors while the ones composing the background have different colors. In this paper, a web application written in javascript is presented, that implements a digital Ishihara-like test for pre-school aged children. Instead numbers or letters, It can transform any binary image representing animal shapes, or any other child-friendly shape, into an Ishihara-like image. This digital plate is not static. The operator can increment the dot density to improve the quality of the shape contour and the entire plate can be redrawn with different dot sizes/colors chosen randomly according to the color pattern of the test. Separate controls for brightness and saturation are implemented to calibrate the chromatic aspect of the background and foreground dots.

A teledentistry system for the second opinion.

In this paper we present a Teledentistry system aimed to the Second Opinion task. It make use of a particular camera called intra-oral camera, also called dental camera, in order to perform the photo shooting and real-time video of the inner part of the mouth. The pictures acquired by the Operator with such a device are sent to the Oral Medicine Expert (OME) by means of a current File Transfer Protocol (FTP) service and the real-time video is channeled into a video streaming thanks to the VideoLan client/server (VLC) application. It is composed by a HTML5 web-pages generated by PHP and allows to perform the Second Opinion both when Operator and OME are logged and when one of them is offline.

Automatic skull stripping in MRI based on morphological filters and fuzzy c-means segmentation.

In this paper a new automatic skull stripping method for T1-weighted MR image of human brain is presented. Skull stripping is a process that allows to separate the brain from the rest of tissues. The proposed method is based on a 2D brain extraction making use of fuzzy c-means segmentation and morphological operators applied on transversal slices. The approach is extended to the 3D case, taking into account the result obtained from the preceding slice to solve the organ splitting problem. The proposed approach is compared with BET (Brain Extraction Tool) implemented in MRIcro software.

Pervasive access to MRI bias artifact suppression service on a grid.

Bias artifact corrupts MRIs in such a way that the image is afflicted by illumination variations. Some of the authors proposed the exponential entropy-driven homomorphic unsharp masking ( E(2)D-HUM) algorithm that corrects this artifact without any a priori hypothesis about the tissues or the MRI modality. Moreover, E(2)D-HUM does not care about the body part under examination and does not require any particular training task. People who want to use this algorithm, which is Matlab-based, have to set their own computers in order to execute it. Furthermore, they have to be Matlab-skilled to exploit all the features of the algorithm. In this paper, we propose to make such algorithm available as a service on a grid infrastructure, so that people can use it almost from everywhere, in a pervasive fashion, by means of a suitable user interface running on smartphones. The proposed solution allows physicians to use the E(2)D-HUM algorithm (or any other kind of algorithm, given that it is available as a service on the grid), being it remotely executed somewhere in the grid, and the results are sent back to the user's device. This way, physicians do not need to be aware of how to use Matlab to process their images. The pervasive service provision for medical image enhancement is presented, along with some experimental results obtained using smartphones connected to an existing Globus-based grid infrastructure.

Blood vessels and feature points detection on retinal images.

In this paper we present a method for the automatic extraction of blood vessels from retinal images, while capturing points of intersection/overlap and endpoints of the vascular tree. The algorithm performance is evaluated through a comparison with handmade segmented images available on the STARE project database (STructured Analysis of the REtina). The algorithm is performed on the green channel of the RGB triad. The green channel can be used to represent the illumination component. The matched filter is used to enhance vessels w.r.t. the background. The separation between vessels and background is accomplished by a threshold operator based on gaussian probability density function. The length filtering removes pixels and isolated segments from the resulting image. Finally endpoints, intersections and overlapping vessels are extracted.

A Gabor-based technique for bias removal in MR images.

Magnetic Resonance images are often characterized by irregularly displaced luminance fluctuations that are called bias artifact. This disturb is due to a drop in signal intensity caused by the distance between imaged sample and receiver coil. An original approach to bias removal in Magnetic Resonance images is presented, which is based on the use of Gabor filter to extract the artifact. The proposed technique restores the image using a correction model, which is derived from the attenuation of signal diffusion across the tissues. No hypotheses are made about the structure of the tissues under investigation and the used MR spectrum. The approach is presented in detail, and extensive experimental results are reported along with a comparison with other popular techniques for bias removal.

Illumination correction on MR images.

OBJECTIVE: An important artifact corrupting Magnetic Resonance Images is the rf inhomogeneity, also called bias artifact. This anomaly produces an abnormal illumination fluctuation on the image, due to variations of the device magnetic field. This artifact is particularly strong on images acquired with a device specialized on upper and lower limbs due to their coil configuration. A method based on homomorphic filtering aimed to suppress this artifact was proposed by Guillemaud. This filter has two faults: it doesn't provide an indication about the cutoff frequency (cf) and introduces another illumination artifact on the edges of the foreground. This work is an improvement to this method because it resolves both problems. METHODS: The experimental setup has been performed on knee images obtained by 5 volunteers and acquired through an Artoscan device using the following parameters: Spin Echo sequence, Repetition time: 980 ms, Echo time: 26 ms, Slice thickness: 4 mm, Flip Angle: 90 degrees . RESULTS: Two specialists in orthoptics evaluated the results of the proposed approach by examining the restored images and validating the results produced by the filter. A quantitative evaluation has been performed on a manually segmented restored image using the coefficient of variation (cv) measure. CONCLUSIONS: Following the specialists qualitative evaluation, the illuminance of upper and lower peripheral zones results to be enhanced; a loose of contrast can be noted only in few cases. The Bias image exhibits an artifact focused usually on the central part of the foreground. The quantitative evaluation based on cv shows that this index is lowered for all the segmented regions with respect to the original value. The method is automatic and doesn't require any hypothesis on the tissues. A manual version of the algorithm can be also implemented allowing the physician to choose the preferred cf. In this case the value selected by the method can be considered as a default value.

Three-dimensional geometrical models of the liver.

In this work we used a virtual approach to study the human liver by three-dimensional geometrical models. We built the models through computer aided geometric modelling techniques starting from pictures taken during both real dissections and diagnostic medical imaging. The results show in a complete modular synthesis and with a schematic iconology the structural organization of this organ in a logic sequence of layers and topographic and spatial relationship among its components. This approach represents an amazing support to clinical anatomy for teaching and research.

Morphological exponential entropy driven-HUM.

This paper presents an improvement to the exponential entropy driven-homomorphic unsharp masking (E(2)D-HUM) algorithm devoted to illumination artifact suppression on magnetic resonance images. E(2)D-HUM requires a segmentation step to remove dark regions in the foreground whose intensity is comparable with background, because strong edges produce streak artifacts on the tissues. This new version of the algorithm keeps the same good properties of E(2)D-HUM without a segmentation phase, whose parameters should be chosen in relation to the image.

Exponential Entropy Driven HUM on Knee MR Images.

A very important artifact corrupting Magnetic Resonance Images is the RF inhomogeneity. This kind of artifact generates variations of illumination which trouble both direct examination by the doctor and segmentation algorithms. Even if homomorphic filtering approaches have been presented in literature, none of them has developed a measure to determine the cut-off frequency. In this work we present a measure based on information theory with a large experimental setup aimed to demonstrate the validity of our approach.

Three-dimensional geometrical models of the inguinal region. Towards a new stereology.

In this work we studied the inguinal-abdominal region and the inguinal canal using three-dimensional geometrical models. We built the models through computer aided geometric modeling techniques on the basis of observations during real dissections, operations and diagnostic medical imaging. The obtained models show in a complete modular synthesis and with a schematic iconology the structural organization of the anatomical districts in a logic sequence of layers and topographic and spatial relationships among its components. The models represent an amazing support to anatomy and clinical anatomy for teaching and research purposes on organogenesis, surgery and diagnosis.