ECG Marker Evaluation for the Machine-Learning-Based Classification of Acute and Chronic Phases of Trypanosoma cruzi Infection in a Murine Model.

Chagas disease (CD) is a neglected parasitic disease caused by the protozoan Trypanosoma cruzi (T. cruzi). The disease has two clinical phases: acute and chronic. In the acute phase, the parasite circulates in the blood. The infection can be asymptomatic or can cause unspecific clinical symptoms. During the chronic phase, the infection can cause electrical conduction abnormalities and progress to cardiac failure. The use of an electrocardiogram (ECG) has been a methodology for diagnosing and monitoring CD, but it is necessary to study the ECG signals to better understand the behavior of the disease. The aim of this study is to analyze different ECG markers using machine-learning-based algorithms for the classification of the acute and chronic phases of T. cruzi infection in a murine experimental model. The presented methodology includes a statistical analysis of control vs. infected models in both phases, followed by an automatic selection of ECG descriptors and the implementation of several machine learning algorithms for the automatic classification of control vs. infected mice in acute and/or chronic phases (binomial classification), as well as a multiclass classification strategy (control vs. the acute group vs. the chronic group). Feature selection analysis showed that P wave duration, R and P wave voltages, and the QRS complex are some of the most important descriptors. The classifiers showed good results in detecting the acute phase of infection (with an accuracy of 87.5%), as well as in multiclass classification (control vs. the acute group vs. the chronic group), with an accuracy of 91.3%. These results suggest that it is possible to detect infection at different phases, which can help in experimental and clinical studies of CD.

Estimating Echocardiographic Myocardial Strain of Left Ventricle with Deep Learning.

The global longitudinal strain of the myocardial tissue has been shown to be a better indicator of cardiac pathologies in the subclinical stage than other indices, such as the ejection fraction. This article presents a new deep learning approach for strain estimation in 2D echocardiograms. The proposed method improves the performance of the state of the art without losing stability with noisy echocardiograms and achieved an average end point error of 0.14 ± 0.17 pixels in the estimation of the optical flow in the myocardium and an error of 1.34 ± 2.34 % in the estimation of the global longitudinal strain indicator when evaluated in a synthetic echocardiographic dataset. Further research will validate the proposed method by a clinical in-vivo dataset. Clinical relevance- This paper presents a method to estimate the global longitudinal strain index in noisy echocardiograms, which promises to be a better indicator of cardiac pathologies in the subclinical stage than other indices such as the ejection fraction.

Neuromorphometry of primary brain tumors by magnetic resonance imaging.

Magnetic resonance imaging is a technique for the diagnosis and classification of brain tumors. Discrete compactness is a morphological feature of two-dimensional and three-dimensional objects. This measure determines the compactness of a discretized object depending on the sum of the areas of the connected voxels and has been used for understanding the morphology of nonbrain tumors. We hypothesized that regarding brain tumors, we may improve the malignancy grade classification. We analyzed the values in 20 patients with different subtypes of primary brain tumors: astrocytoma, oligodendroglioma, and glioblastoma multiforme subdivided into the contrast-enhanced and the necrotic tumor regions. The preliminary results show an inverse relationship between the compactness value and the malignancy grade of gliomas. Astrocytomas exhibit a mean of [Formula: see text], whereas oligodendrogliomas exhibit a mean of [Formula: see text]. In contrast, the contrast-enhanced region of the glioblastoma presented a mean of [Formula: see text], and the necrotic region presented a mean of [Formula: see text]. However, the volume and area of the enclosing surface did not show a relationship with the malignancy grade of the gliomas. Discrete compactness appears to be a stable characteristic between primary brain tumors of different malignancy grades, because similar values were obtained from different patients with the same type of tumor.

Volumetric electromagnetic phase-shift spectroscopy of brain edema and hematoma.

Motivated by the need of poor and rural Mexico, where the population has limited access to advanced medical technology and services, we have developed a new paradigm for medical diagnostic based on the technology of "Volumetric Electromagnetic Phase Shift Spectroscopy" (VEPS), as an inexpensive partial substitute to medical imaging. VEPS, can detect changes in tissue properties inside the body through non-contact, multi-frequency electromagnetic measurements from the exterior of the body, and thereby provide rapid and inexpensive diagnostics in a way that is amenable for use in economically disadvantaged parts of the world. We describe the technology and report results from a limited pilot study with 46 healthy volunteers and eight patients with CT radiology confirmed brain edema and brain hematoma. Data analysis with a non-parametric statistical Mann-Whitney U test, shows that in the frequency range of from 26 MHz to 39 MHz, VEPS can distinguish non-invasively and without contact, with a statistical significance of p<0.05, between healthy subjects and those with a medical conditions in the brain. In the frequency range of between 153 MHz to 166 MHz it can distinguish with a statistical significance of p<0.05 between subjects with brain edema and those with a hematoma in the brain. A classifier build from measurements in these two frequency ranges can provide instantaneous diagnostic of the medical condition of the brain of a patient, from a single set of measurements. While this is a small-scale pilot study, it illustrates the potential of VEPS to change the paradigm of medical diagnostic of brain injury through a VEPS classifier-based technology. Obviously substantially larger-scale studies are needed to verify and expand on the findings in this small pilot study.

Computer assisted biopsy of breast tumors.

In this paper we report our preliminary results of the development of a computer assisted system for breast biopsy. The system is based on tracked ultrasound images of the breast. A three dimensional ultrasound volume is constructed from a set of tracked B-scan images acquired with a calibrated probe. The system has been designed to assist a radiologist during breast biopsy, and also as a training system for radiology residents. A semiautomatic classification algorithm was implemented to assist the user with the annotation of the tumor on an ultrasound volume. We report the development of the system prototype, tested on a physical phantom of a breast with a tumor, made of polivinil alcohol.

Prediction of infarct growth based on apparent diffusion coefficients: penumbral assessment without intravenous contrast material.

PURPOSE: To compare predicted and final infarct lesion volumes determined by processing apparent diffusion coefficient (ADC) maps derived at admission diffusion-weighted (DW) magnetic resonance (MR) imaging in patients with acute stroke and to verify that predicted areas of infarct growth reflect at-risk penumbral regions based on recanalization status. MATERIALS AND METHODS: The French legislation waived the requirement for informed patient consent for the described research, which was based on patient medical files. However, patients and/or their relatives were informed that they could decline to participate in the research. Authors tested a semiautomated proprietary image analysis procedure in 98 patients with middle cerebral artery (MCA) stroke by modeling infarct growth on DW imaging-derived ADC maps. Predicted infarct growth (PIG) areas and predicted infarct volumes were correlated with final observed data. In addition, the effect of MCA recanalization on the correlation between predicted and observed infarct growth volumes was qualitatively assessed. RESULTS: Predicted and final infarct volumes (rho = 0.828; 95% confidence interval [CI]: 0.753, 0.882; P < .0001) and infarct growth volumes (rho = 0.506; 95% CI: 0.342, 0.640; P < .0001) were significantly correlated. Visual comparative examination revealed satisfactory qualitative consistency between predicted and follow-up lesion masks. In patients without MCA recanalization, PIG did not differ significantly from final observed infarct growth (median PIG obtained with 0.93 ADC ratio cutoff [PIG(ratio)] of 27.1 cm(3) vs median infarct growth of 19.8 cm(3), P = .17). MCA recanalization revealed an overestimation of PIG (median PIG(ratio) of 24.8 cm(3) vs median infarct growth of 12 cm(3), P = .005), suggesting that the PIG area was part of ischemic penumbra. CONCLUSION: Data show the feasibility of identifying at-risk ischemic tissue in patients with acute MCA stroke by using semiautomated analysis of ADC maps derived at DW imaging, without intravenous contrast material-enhanced perfusion-weighted imaging.

Robust nonparametric segmentation of infarct lesion from diffusion-weighted MR images.

Magnetic Resonance Imaging (MRI) is increasingly used for the diagnosis and monitoring of neurological disorders. In particular Diffusion-Weighted MRI (DWI) is highly sensitive in detecting early cerebral ischemic changes in acute stroke. Cerebral infarction lesion segmentation from DWI is accomplished in this work by applying nonparametric density estimation. The quality of the class boundaries is improved by including an edge confidence map, that is the confidence of truly being in the presence of a border between adjacent regions. The adjacency graph, that is constructed with the label regions, is analyzed and pruned to merge adjacent regions. The method was applied to real images, keeping all parameters constant throughout the process for each data set. The combination of region segmentation and edge detection proved to be a robust automatic technique of segmentation from DWI images of cerebral infarction regions in acute ischemic stroke. In a comparison with the reference infarct lesions segmentation, the automatic segmentation presented a significant correlation (r=0.935), and an average Tanimoto index of 0.538.