e-CoVig: A Novel mHealth System for Remote Monitoring of Symptoms in COVID-19

In 2019, a new virus, SARS-CoV-2, responsible for the COVID-19 disease, was discovered. Asymptomatic and mildly symptomatic patients were forced to quarantine and closely monitor their symptoms and vital signs, most of the time at home. This paper describes e-CoVig, a novel mHealth application, developed as an alternative to the current monitoring paradigm, where the patients are followed up by direct phone contact. The e-CoVig provides a set of functionalities for remote reporting of symptoms, vital signs, and other clinical information to the health services taking care of these patients. The application is designed to register and transmit the heart rate, blood oxygen saturation (SpO2), body temperature, respiration, and cough. The system features a mobile application, a web/cloud platform, and a low-cost specific device to acquire the temperature and SpO2. The architecture of the system is flexible and can be configured for different operation conditions. Current commercial devices, such as oximeters and thermometers, can also be used and read using the optical character recognition (OCR) functionality of the system. The data acquired at the mobile application are sent automatically to the web/cloud application and made available in real-time to the medical staff, enabling the follow-up of several users simultaneously without the need for time consuming phone call interactions. The system was already tested for its feasibility and a preliminary deployment was performed on a nursing home showing promising results.

Treatment of early-stage mycosis fungoides: results from the PROspective Cutaneous Lymphoma International Prognostic Index (PROCLIPI) study

BACKGROUND

The PROspective Cutaneous Lymphoma International Prognostic Index (PROCLIPI) study is a prospective analysis of an international database. Here we examine front-line treatments and quality of life (QoL) in patients with newly diagnosed mycosis fungoides (MF).

OBJECTIVES

To identify (i) differences in first-line approaches according to tumour-nodes-metastasis-blood (TNMB) staging; (ii) parameters related to a first-line systemic approach and (iii) response rates and QoL measures.

METHODS

In total, 395 newly diagnosed patients with early-stage MF (stage IA-IIA) were recruited from 41 centres in 17 countries between 1 January 2015 and 31 December 2018 following central clinicopathological review.

RESULTS

The most common first-line therapy was skin-directed therapy (SDT) (322 cases, 81·5%), while a smaller percentage (44 cases, 11·1%) received systemic therapy. Expectant observation was used in 7·3%. In univariate analysis, the use of systemic therapy was significantly associated with higher clinical stage (IA, 6%; IB, 14%; IIA, 20%; IA-IB vs. IIA, P < 0·001), presence of plaques (T1a/T2a, 5%; T1b/T2b, 17%; P < 0·001), higher modified Severity Weighted Assessment Tool (> 10, 15%; ≤ 10, 7%; P = 0·01) and folliculotropic MF (FMF) (24% vs. 12%, P = 0·001). Multivariate analysis demonstrated significant associations with the presence of plaques (T1b/T2b vs. T1a/T2a, odds ratio 3·07) and FMF (odds ratio 2·83). The overall response rate (ORR) to first-line SDT was 73%, while the ORR to first-line systemic treatments was lower (57%) (P = 0·027). Health-related QoL improved significantly both in patients with responsive disease and in those with stable disease.

CONCLUSIONS

Disease characteristics such as presence of plaques and FMF influence physician treatment choices, and SDT was superior to systemic therapy even in patients with such disease characteristics. Consequently, future treatment guidelines for early-stage MF need to address these issues.

Atomic force microscopy and graph analysis to study the P-cadherin/SFK mechanotransduction signalling in breast cancer cells

Physical forces mediated by cell-cell adhesion molecules, as cadherins, play a crucial role in preserving normal tissue architecture. Accordingly, altered cadherins' expression has been documented as a common event during cancer progression. However, in most studies, no data exist linking pro-tumorigenic signaling and variations in the mechanical balance mediated by adhesive forces. In breast cancer, P-cadherin overexpression increases in vivo tumorigenic ability, as well as in vitro cell invasion, by activating Src family kinase (SFK) signalling. However, it is not known how P-cadherin and SFK activation impact cell-cell biomechanical properties. In the present work, using atomic force microscopy (AFM) images, cell stiffness and cell-cell adhesion measurements, and undirected graph analysis based on microscopic images, we have demonstrated that P-cadherin overexpression promotes significant alterations in cell's morphology, by decreasing cellular height and increasing its area. It also affects biomechanical properties, by decreasing cell-cell adhesion and cell stiffness. Furthermore, cellular network analysis showed alterations in intercellular organization, which is associated with cell-cell adhesion dysfunction, destabilization of an E-cadherin/p120ctn membrane complex and increased cell invasion. Remarkably, inhibition of SFK signaling, using dasatinib, reverted the pathogenic P-cadherin induced effects by increasing cell's height, cell-cell adhesion and cell stiffness, and generating more compact epithelial aggregates, as quantified by intercellular network analysis. In conclusion, P-cadherin/SFK signalling induces topological, morphological and biomechanical cell-cell alterations, which are associated with more invasive breast cancer cells. These effects could be further reverted by dasatinib treatment, demonstrating the applicability of AFM and cell network diagrams for measuring the epithelial biomechanical properties and structural organization.

Automated stratification of liver disease in ultrasound: An online accurate feature classification paradigm

PURPOSE

Fatty liver disease (FLD) is one of the most common diseases in liver. Early detection can improve the prognosis considerably. Using ultrasound for FLD detection is highly desirable due to its non-radiation nature, low cost and easy use. However, the results can be slow and ambiguous due to manual detection. The lack of computer trained systems leads to low image quality and inefficient disease classification. Thus, the current study proposes novel, accurate and reliable detection system for the FLD using computer-based training system.

MATERIALS AND METHODS

One hundred twenty-four ultrasound sample images were selected retrospectively from a database of 62 patients consisting of normal and cancerous. The proposed training system was generated offline parameters using training liver image database. The classifier applied transformation parameters to an online system in order to facilitate real-time detection during the ultrasound scan. The system utilized six sets of features (a total of 128 features), namely Haralick, basic geometric, Fourier transform, discrete cosine transform, Gupta transform and Gabor transform. These features were extracted for both offline training and online testing. Levenberg-Marquardt back propagation network (BPN) classifier was used to classify the liver disease into normal and abnormal categories.

RESULTS

Random partitioning approach was adapted to evaluate the classifier performance and compute its accuracy. Utilizing all the six sets of 128 features, the computer aided diagnosis (CAD) system achieved classification accuracy of 97.58%. Furthermore, the four performance metrics consisting of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) realized 98.08%, 97.22%, 96.23%, and 98.59%, respectively.

CONCLUSION

The proposed system was successfully able to detect and classify the FLD. Furthermore, the proposed system was benchmarked against previous methods. The comparison established an advanced set of features in the Levenberg-Marquardt back propagation network reports a significant improvement compared to the existing techniques.

Understanding symptomatology of atherosclerotic plaque by image-based tissue characterization

Characterization of carotid atherosclerosis and classification into either symptomatic or asymptomatic is crucial in terms of diagnosis and treatment planning for a range of cardiovascular diseases. This paper presents a computer-aided diagnosis (CAD) system (Atheromatic) that analyzes ultrasound images and classifies them into symptomatic and asymptomatic. The classification result is based on a combination of discrete wavelet transform, higher order spectra (HOS) and textural features. In this study, we compare support vector machine (SVM) classifiers with different kernels. The classifier with a radial basis function (RBF) kernel achieved an average accuracy of 91.7% as well as a sensitivity of 97%, and specificity of 80%. Thus, it is evident that the selected features and the classifier combination can efficiently categorize plaques into symptomatic and asymptomatic classes. Moreover, a novel symptomatic asymptomatic carotid index (SACI), which is an integrated index that is based on the significant features, has been proposed in this work. Each analyzed ultrasound image yields on SACI number. A high SACI value indicates that the image shows symptomatic and low value indicates asymptomatic plaques. We hope this SACI can support vascular surgeons during routine screening for asymptomatic plaques.

Geometric correction of deformed chromosomes for automatic Karyotyping

Automatic Karyotyping is the process of classifying chromosomes from an unordered karyogram into their respective classes to create an ordered karyogram. Automatic karyotyping algorithms typically perform geometrical correction of deformed chromosomes for feature extraction; these features are used by classifier algorithms for classifying the chromosomes. Karyograms of bone marrow cells are known to have poor image quality. An example of such karyograms is the Lisbon-K(1) (LK(1)) dataset that is used in our work. Thus, to correct the geometrical deformation of chromosomes from LK(1), a robust method to obtain the medial axis of the chromosome was necessary. To address this problem, we developed an algorithm that uses the seed points to make a primary prediction. Subsequently, the algorithm computes the distance of boundary from the predicted point, and the gradients at algorithm-specified points on the boundary to compute two auxiliary predictions. Primary prediction is then corrected using auxiliary predictions, and a final prediction is obtained to be included in the seed region. A medial axis is obtained this way, which is further used for geometrical correction of the chromosomes. This algorithm was found capable of correcting geometrical deformations in even highly distorted chromosomes with forked ends.

Atheromatic™: symptomatic vs. asymptomatic classification of carotid ultrasound plaque using a combination of HOS, DWT & texture

Quantitative characterization of carotid atherosclerosis and classification into either symptomatic or asymptomatic is crucial in terms of diagnosis and treatment planning for a range of cardiovascular diseases. This paper presents a computer-aided diagnosis (CAD) system (Atheromatic™, patented technology from Biomedical Technologies, Inc., CA, USA) which analyzes ultrasound images and classifies them into symptomatic and asymptomatic. The classification result is based on a combination of discrete wavelet transform, higher order spectra and textural features. In this study, we compare support vector machine (SVM) classifiers with different kernels. The classifier with a radial basis function (RBF) kernel achieved an accuracy of 91.7% as well as a sensitivity of 97%, and specificity of 80%. Encouraged by this result, we feel that these features can be used to identify the plaque tissue type. Therefore, we propose an integrated index, a unique number called symptomatic asymptomatic carotid index (SACI) to discriminate symptomatic and asymptomatic carotid ultrasound images. We hope this SACI can be used as an adjunct tool by the vascular surgeons for daily screening.

ATP consumption and neural electrical activity: a physiological model for brain imaging

The relation between neural electrical activity and oxygen consumption is the key issue in almost all brain image modalities based on perfusion. Despite the large amount of physiological information available in the literature about the processes involved in neural activation, a practical, tractable and simultaneously accurate mathematical model to describe this relation is needed.

Classifier-assisted metric for chromosome pairing

Cytogenetics plays a central role in the detection of chromosomal abnormalities and in the diagnosis of genetic diseases. A karyogram is an image representation of human chromosomes arranged in order of decreasing size and paired in 23 classes. In this paper we propose an approach to automatically pair the chromosomes into a karyogram, using the information obtained in a rough SVM-based classification step, to help the pairing process mainly based on similarity metrics between the chromosomes. Using a set of geometric and band pattern features extracted from the chromosome images, the algorithm is formulated on a Bayesian framework, combining the similarity metric with the results from the classifier. The solution is obtained solving a mixed integer program. Two datasets with contrasting quality levels and 836 chromosomes each were used to test and validate the algorithm. Relevant improvements with respect to the algorithm described by the authors in [1] were obtained with average paring rates above 92%, close to the rates obtained by human operators.

Robust band profile extraction using constrained nonparametric machine-learning technique

A typical characteristic of images of bone marrow cells taken during mitosis is poor quality. This renders the task of extraction of accurate band profile, representative of intensity distribution over each chromosome, more challenging. A robust method is hence required to tackle this problem. An algorithm was thus developed, which estimates a single-line medial axis, the basis for computation of band profile. Medial axis was generated by computing a final prediction, using primary and secondary predictions obtained by a nonparametric machine learning algorithm trained with data from chromosome's skeleton, and geometrical properties of medial axis, respectively. Experiments were performed using the LK(1) dataset. The algorithm was found capable of estimating a satisfactory single-line medial axis. Band profile obtained was found to be a good representation of intensity levels in different regions of chromosomes. Additionally, this algorithm is robust in terms of growing a very small seed region into desired medial axis and also handling highly irregular chromosomes.

Denoising of LSFCM images with compensation for the photoblinking/photobleaching effects

Fluorescence confocal microscopy images present a low signal to noise ratio and a time intensity decay due to the so called photoblinking and photobleaching effects. These effects, together with the Poisson multiplicative noise that corrupts the images, make long time biological observation processes very difficult.

Bayesian optimization of perfusion and transit time estimation in PASL-MRI

Pulsed Arterial Spin Labeling (PASL) techniques potentially allow the absolute, non-invasive quantification of brain perfusion and arterial transit time. This can be achieved by fitting a kinetic model to the data acquired at a number of inversion time points (TI). The intrinsically low SNR of PASL data, together with the uncertainty in the model parameters, can hinder the estimation of the parameters of interest. Here, a two-compartment kinetic model is used to estimate perfusion and transit time, based on a Maximum a Posteriori (MAP) criterion. A priori information concerning the physiological variation of the multiple model parameters is used to guide the solution. Monte Carlo simulations are performed to compare the accuracy of our proposed Bayesian estimation method with a conventional Least Squares (LS) approach, using four different sets of TI points. Each set is obtained either with a uniform distribution or an optimal sampling strategy designed based on the same MAP criterion. We show that the estimation errors are minimized when our proposed Bayesian estimation method is employed in combination with an optimal set of sampling points. In conclusion, our results indicate that PASL perfusion and transit time measurements would benefit from a Bayesian approach for the optimization of both the sampling strategy and the estimation algorithm, whereby prior information on the parameters is used.

Three-dimensional labeling of vulnerable regions in carotid plaques using Graph-Cuts

Carotid plaque morphology assessed through high resolution ultrasound is nowadays an important diagnostic approach in the evaluation of stroke risk. The literature describes several methods for the characterization of plaque echogenicity and texture by using overall statistics across the carotid plaques. However, this averaged indicators may not be enough for a correct diagnosis of plaque instability. Therefore, it has been suggested that a local characterization of plaque morphology, where the extension and location of vulnerable regions inside the plaques is taken into account, could lead to significant improvements in the assessment of stroke risk. In this paper, a new method to characterize the morphology of carotid plaques in three-dimensions (3D) is proposed. The local characterization of carotid plaques is performed by using a 3D Graph-Cuts (GraphC) robust labeling method. This methodology allows to binary segment the data by minimizing an energy function that uses spatial correlation among neighboring pixels/voxels in order to remove small misclassified regions. The method is efficient since it is able to find the global minimum of the energy function in a very short time interval. Results show that this labeling procedure is less noisier and favors clustering, being more meaningful from a clinical point of view than the one obtained with simple thresholding. This paper shows that the use of GraphC may improve the 3D morphological characterization of carotid plaques, namely by providing a more appropriate identification of unstable foci inside the plaque.

Automatic chromosome pairing using mutual information

Cytogenetics is a key tool in the detection of acquired chromosomal abnormalities and in the diagnosis of genetic diseases such as leukemia. The karyotyping is a set of procedures, in the scope of the cytogenetics, that produces a visual representation of the 46 chromosomes (called karyogram), paired and arranged in decreasing order of size. The pairing procedure aims to identify all pairs of homologous chromosomes.The pairing criterion is based on dimensional, morphological,and textural features similarity. This process is time consuming when performed manually, and demanding from a technical point of view. An automatic pairing algorithm would thus bring benefits, but it remains an open problem to date.In this paper a new strategy for automatic pairing of homologous chromosomes is proposed. Besides the traditional features described in the literature, the Mutual Information (MI) is used to discriminate chromosome textural differences. A supervised non-linear classifier is trained by using manual classifications provided by expert technicians, combining the different features computed from each pair.Simulations using 836 real chromosome images, obtained with a Leica Optical Microscope DM 2500, in a leave-one-out cross validation fashion, were performed for training and testing the algorithm.Promising and relevant results were found, despite the poor quality of the original chromosome images, contrasting with state-of the-art algorithms and datasets found in the literature.

Convex ultrasound image reconstruction with log-Euclidean priors

Image reconstruction from noisy and incomplete observations is usually an ill-posed problem. A Bayesian framework may be adopted do deal with this such inverse task by well posing the reconstruction problem. In this approach, the ill poseness nature of the reconstruction is removed by minimizing a two-term energy function. The first term pushes the solution toward the data and the second regularizes the solution. A Bayesian algorithm for ultrasound image reconstruction and de-noising is proposed where an edge preserving prior is used to reduce the smoothing effect at the transitions. The prior distribution is based on log-Euclidean potential functions that are particular suitable in reconstruction problems under the constraint of positivity, that is, when the unknowns to be estimated should be positive, which is the case, where the noisy observations are modeled by a Rayleigh distribution. In this paper, the reconstruction procedure is formulated as the optimization of a convex function and a Newton method is adopted to obtain the minimizer. This strategy guarantees a convergence to the global minimum in a small number of iterations. Experimental results, using synthetic and real medical images are shown. The proposed method produces images where speckle noise is effectively suppressed and important clinical details (organ and tissue transitions) are preserved.

Modeling log-compressed ultrasound images for radio frequency signal recovery

This paper presents an algorithm for recovering the radio frequency (RF) signal provided by the ultrasound probe from the log-compressed ultrasound images displayed in ultrasound equipment. Commercial ecographs perform nonlinear image compression to reduce the dynamic range of the Ultrasound (US) signal in order to improve image visualization. Moreover, the clinician may adjust other parameters, such as brightness, gain and contrast, to improve image quality of a given anatomical detail. These operations significantly change the statistical distribution of the original RF raw signal, which is assumed, based on physical considerations on the signal formation process, to be Rayleigh distributed. Therefore, the image pixels are no longer Rayleigh distributed and the RF signal is not usually available in the common ultrasound equipment. For statistical data processing purposes, more important than having "good looking" images, it is important to have realistic models to describe the data. In this paper, a nonlinear compression parametric function is used to model the pre-processed image in order to recover the original RF image as well the contrast and brightness parameters. Tests using synthetic and real data and statistical measures such as the Kolmogorov-Smirnov and Kullback-Leibler divergences are used to assess the results. It is shown that the proposed estimation model clearly represents better the observed data than by taking the general assumption of the data being modeled by a Rayleigh distribution.

Total variation with automatic hyper-parameter estimation

Medical diagnosis is often hampered by the quality of the images. This happens in a wide range of image modalities. Image noise reduction is a crucial step, however difficult to be accomplished. Bayesian algorithms have been commonly used with success, namely with additive white Gaussian noise (AWGN) model. In fact, the noise corrupting some of the most used medical imaging modalities is not additive neither Gaussian but multiplicative described by Poisson or Rayleigh distributions. This paper proposes a unified framework with automatic hyper parameters estimation. The proposed framework deals with AWGN but also with both Poisson and Rayleigh distributions. The algorithm proposed herein, is based on a maximum a posteriori (MAP) criterion with the edge preserving prior based on the total variation (TV), which avoids the distortion of relevant anatomical details. The denoising technique is performed via single parametric iterative scheme parameterized for each noise model considered. Tests with real data from several medical imaging modalities testify the performance of the algorithm.

[Broncho-splenic fistula caused by hydatidosis]

To our knowledge, this is the first case report of a broncho-splenic fistula of hydatid origin. We discuss the clinical, radiological and therapeutic aspects of this rare complication of hydatid disease.

[Imaging of aortic aneurysms and dissections. Part 2. Aortic dissections]

The authors review the main imaging findings of aneurysms and dissections of the aorta. Based on the experience and imaging techniques available in their institution (Conventional Radiology, Sonography and Computed Tomography), they try to make a decision tree of radiologic procedures in ambulatory and emergency patients. Nowadays, in the majority of situations, these techniques allow the Radiologist to make the correct diagnosis, detect the complications and evaluate the patients in the pre and post-surgical phases. Other imaging techniques not available in their institution, such as Magnetic Resonance Imaging and Aortography, will be mentioned in view of their usefulness in some specific situations.

[Imaging of aneurysms and dissections of the aorta. Part 1--Aneurysms of the abdominal aorta]

The authors review the main imaging findings of aneurysms and dissections of the aorta. Based on the experience and imaging techniques available in their institution (Conventional Radiology, Sonography and Computed Tomography), they try to make a decision tree of radiologic procedures in ambulatory and emergency patients. Nowadays, in the majority of situations, these techniques allow the Radiologist to make the correct diagnosis, detect the complications and evaluate the patients in the pre and post-surgical phases. Other imaging techniques not available in their institution, such as Magnetic Resonance Imaging and Aortography, will be mentioned in view of their usefulness in some specific situations.

Posterior fossa hemangioblastomas: MR imaging

The magnetic resonance (MR) imaging findings of 18 surgically proved posterior fossa hemangioblastomas (15 patients) were retrospectively analyzed and correlated with computed tomographic (ten patients) and angiographic (eight patients) findings. Thirteen tumors were located in the cerebellar hemisphere, three in the vermis, and two in the medulla with associated syrinxes. Three patients had von Hippel-Lindau disease, two of whom had multiple cerebellar hemangioblastomas. Seven hemangioblastomas appeared as solid tumors, six as solid masses with central cysts, and five as cysts with mural nodules. Abnormal tumor vessels, with characteristic signal void, were demonstrated in 13 tumors. Associated hemorrhage was present in four tumors. Although angiography is usually required for the diagnosis and preoperative assessment of this tumor, MR imaging demonstration of a posterior fossa mass with abnormal vessels should suggest the diagnosis of hemangioblastoma. Moreover, the combination of a peripheral posterior fossa cyst with a mural nodule supplied by enlarged vessels may be pathognomonic.