PosterPDF Available

Feasibility study of a portable kit for Chagas-Mazza disease diagnosis and data centralization

July 2012

DOI:10.13140/2.1.4914.0487

Authors:

Emmanuel Luján

Massachusetts Institute of Technology

The goal of this project is to study the development of a portable kit for the diagnosis of Chagas-Mazza disease for use in remote areas distant from health centers. The portable kit consists in two devices: a detection device implemented in a board with a field programmable gate array (FPGA) for screening blood samples, and a diagnostic device implemented in a notebook or in a smart cell phone, for patient data storage and final diagnosis processing. The detection device is used for screening blood samples to detect T. cruzi parasites rapid motion. The diagnostic device is used for enter patient data, processing it using a fuzzy logic based software and diagnosis delivering. A further step in the diagnostic device is to transform data into DICOM format files and transmit it to a PACS server for further analysis by specialists. The large amount of blood sample images underlying T. cruzi rapid motion screening requires the use of High Performance Computing (HPC) for obtaining faster diagnosis. Chagas-Mazza disease is not curable in its chronic (last) stage, but can be treated, this is the reason why early diagnosis is a matter of utmost importance.

Content uploaded by Emmanuel Luján

Content may be subject to copyright.

A preview of the PDF is not available

ResearchGate has not been able to resolve any citations for this publication.

ResearchGate has not been able to resolve any references for this publication.

Article

Temporal Convolutional Neural Networks for Diagnosis from Lab Tests

November 2015

Early diagnosis of treatable diseases is essential for improving healthcare, and many diseases' onsets are predictable from annual lab tests and their temporal trends. We introduce a multi-resolution convolutional neural network for early detection of multiple diseases from irregularly measured sparse lab values. Our novel architecture takes as input both an imputed version of the data and a ... [Show full abstract] binary observation matrix. For imputing the temporal sparse observations, we develop a flexible, fast to train method for differentiable multivariate kernel regression. Our experiments on data from 300K individuals over 8 years, 18 common lab measurements, and 170 diseases show that the temporal signatures learned via convolution are significantly more predictive than baselines commonly used for early disease diagnosis.

Article

Diagnostik duraler Sinus- und zerebraler Venenthrombosen

December 2001 · Clinical Neuroradiology

Armin K. Thron

Background: In dural sinus and cerebral venous thrombosis mortality and morbidity mostly depend on an early diagnosis. This can only be achieved if knowledge exists about suspect findings in routine examinations and about the unspecific early clinical symptoms of the disease, giving rise to a specific examination procedure. Conventional Imaging Techniques: Most cases of dural sinus thrombosis can ... [Show full abstract] be suspected on the basis of CT prior or after administration of contrast medium. In the early phase of the disease diagnosis based on MR imaging without the use of flow-sensitive sequences may face serious problems. Modern Imaging Procedures: Using additional CT angiography, flow-sensitive MR sequences or MR angiography the diagnosis of cerebral venous thrombosis can be established in the majority of suspected cases without the necessity of digital subtraction angiography. The main problem of modern MR imaging is a false positive diagnosis caused by artefacts in MR angiograms.

Article

Serum Proteomics in Biomedical Research: A Systematic Review

April 2013 · Applied Biochemistry and Biotechnology

Proteins that are important indicators of physiological or pathological states may contribute to the early diagnosis of disease, which may provide a basis for identifying the underlying mechanism of disease development. Serum, contains an abundance of proteins, offers an easy and inexpensive approach for disease detection and possesses a high potential to revolutionize the diagnostics. These ... [Show full abstract] differentially expressed proteins in serum have become an important role to monitoring the state for disease. Availability of emerging proteomic techniques gives optimism that serum can eventually be placed as a biomedium for clinical diagnostics. Advancements have benefited biomarker research to the point where serum is now recognized as an excellent diagnostic medium for the detection of disease. Comprehensive proteome of human serum fluid with high accuracy and availability has the potential to open new doors for disease biomarker discovery and for disease diagnostics, providing insights useful for future study. Thus, this review presents an overview of the value of serum as a credible diagnostic tool, and we aim to summarize the proteomic technologies currently used for global analysis of serum proteins and to elaborate on the application of serum proteomics to the discovery of disease biomarkers, and discuss some of the critical challenges and perspectives for this emerging field.

Article

Clinical applications of ultra-high field magnetic resonance imaging in multiple sclerosis

January 2018 · Expert Review of Neurotherapeutics

Introduction: Magnetic resonance imaging (MRI) is of paramount importance for the early diagnosis of multiple sclerosis (MS) and MRI findings are part of the MS diagnostic criteria. There is a growing interest in the use of ultra-high-field strength -7 Tesla- (7T) MRI to investigate, in vivo, the pathological substrate of the disease. Areas covered: An overview of 7T MRI applications in MS ... [Show full abstract] focusing on increased sensitivity for lesion detection, specificity of the central vein sign and better understanding of MS pathophysiology and discuss the implications for disease diagnosis, monitoring and treatment planning. Expert commentary: 7T MRI provides increased signal-to-noise and contrast-to-noise-ratio that allow higher spatial resolution and better detection of anatomical and pathological features. The high spatial resolution reachable at 7T has been a game changer for neuroimaging applications not only in multiple sclerosis but also in epilepsy, brain tumors, dementia, and neuro-psychiatric disorders. Furthermore, the first 7T device has recently been cleared for clinical use by the food and drug administration.

Preprint

Full-text available

Fully Convolutional Neural Network for Semantic Segmentation of Anatomical Structure and Pathologies...

February 2019

Diabetic retinopathy (DR) is the most common form of diabetic eye disease. Retinopathy can affect all diabetic patients and becomes particularly dangerous, increasing the risk of blindness, if it is left untreated. The success rate of its curability solemnly depends on diagnosis at an early stage. The development of automated computer aided disease diagnosis tools could help in faster detection ... [Show full abstract] of symptoms with a wider reach and reasonable cost. This paper proposes a method for the automated segmentation of retinal lesions and optic disk in fundus images using a deep fully convolutional neural network for semantic segmentation. This trainable segmentation pipeline consists of an encoder network, a corresponding decoder network followed by pixel-wise classification to segment microaneurysms, hemorrhages, hard exudates, soft exudates, optic disk from background. The network was trained using Binary cross entropy criterion with Sigmoid as the last layer, while during an additional SoftMax layer was used for boosting response of single class. The performance of the proposed method is evaluated using sensitivity, positive prediction value (PPV) and accuracy as the metrices. Further, the position of the Optic disk is localised using the segmented output map.