149 research outputs found
Sort by
Real-time sleep apnea detection using wavelet packet transform and support vector machines
Get PDFSleep apnea events as obstructive, central, mixed, or hypopnea are characterized by frequent breathing cessations or reduction in upper airflow during sleep. An advanced method for analyzing the patterning of biomedical signals to recognize obstructive sleep apnea and hypopnea is presented. In the aim to extract characteristic parameters, which will be used for classifying the above-stated (obstructive, central, mixed) sleep apnea and hypopnea, the proposed method is based, first, on the analysis of polysomnopraphy signals such as electrocardiogram signal (ECG) and electromyogram (EMG) and then classification of (obstructive, central, mixed) sleep apnea and hypopnea. The analysis is carried out using the wavelet transform technique in order to extract characteristic parameters, whereas classification is carried out by applying the SVM (support vector machine) technique. The obtained results show good recognition rates using characteristic parameters
Detection of muscle fatigue, from statistical methods to software applications
Get PDFBackground: Muscle fatigue has become increasingly present in our daily lives. This is related to lifestyle difficulties. Several methods were proposed in order to detect the muscle fatigue. This report proposes a short review of statistical methods and processing tools extracted from MATLAB software dedicated to detection of fatigue.
Methods: The first part in this study is an application of a useful electronic card named “arduino†for acquiring the electromyography signal (EMG). This latter is the perfect signal to describe fatigue in muscles. The acquisition of data is done in two steps; the first is a simple acquisition representing rest (the subject is relaxed). Then, in the next step, the subject does a series of physical exercises representing moving continuously a handlebar (in order to simulate the work of the tram’s conductor). After obtaining raw data from the acquired signals, we apply statistical methods and some processing tools in order to detect fatigue.
Results: For the statistical method, we apply the spectral density, which is a mathematical tool that represents the various spectral components of a signal and to perform the harmonic analysis. We deduct that 80 microvolt’s is the intensity of getting fatigue (exercise of moving the handlebar). Using processing tools (FFT and STFT techniques), we obtain essential information on the fatigue’s beginning time.
Conclusion: A brief survey of statistical and processing tools to extract fatigue information from an EMG signal was done. A typical example of the importance of detecting fatigue was also illustrated (tram conductor). We aimed for a lot of results from this study, especially because we want to compare techniques. After studying, the STFT technique seem the best
The Rediscovery of the Social Side of Medicine: Philosophy and Value of the International Classification of Functioning, Disability and Health (ICF)
Get PDFoai:ojs2.medtech.ichsmt.org:article/2Medicine is at a risk to slide into a sole repair service for the malfunction of organs. But the patients’ hopes and confidence toward doctors practicing this repair work go far beyond that: after acute medical treatment, many patients suffer from chronic impairments due to the natural course of disease or as a result of medical interventions. Despite the resulting handicaps, patients aim toward participating in family and social life, retaining a workplace, and receiving support to remain a valued member of family and community. Doctors should therefore not only concentrate on the natural science and technological part of medicine but also consider the background of their patients and their involvement in life situations, including environmental and personal factors, as these may influence functioning and disability as facilitators or barriers. Health insurances must organize, finance, and control the achievements of the post-acute treatment process with the goal of participation. Public health must combine and assess individual views to prepare reasonable population-based social, economic, and political decisions. The philosophy and structure of the International Classification of Functioning, Disability and Health (ICF) is supporting this attitude of medicine to complement the International Classification of Diseases (ICD) as a basis for health reports
Biomedical Cyber-Physical Systems in the Light of Database as a Service (DBaaS) Paradigm
Get PDFBackground: A database (DB) to store indexed information about drug delivery, test, and their temporal behavior is paramount in new Biomedical Cyber-Physical Systems (BCPSs). The term Database as a Service (DBaaS) means that a corporation delivers the hardware, software, and other infrastructure required by companies to operate their databases according to their demands instead of keeping an internal data warehouse.
Methods: BCPSs attributes are presented and discussed. One needs to retrieve detailed knowledge reliably to make adequate healthcare treatment decisions. Furthermore, these DBs store, organize, manipulate, and retrieve the necessary data from an ocean of Big Data (BD) associated processes. There are Search Query Language (SQL), and NoSQL DBs.
Results: This work investigates how to retrieve biomedical-related knowledge reliably to make adequate healthcare treatment decisions. Furthermore, Biomedical DBaaSs store, organize, manipulate, and retrieve the necessary data from an ocean of Big Data (BD) associated processes.
Conclusion: A NoSQL DB allows more flexibility with changes while the BCPSs are running, which allows for queries and data handling according to the context and situation. A DBaaS must be adaptive and permit the DB management within an extensive variety of distinctive sources, modalities, dimensionalities, and data handling according to conventional ways
Content-Based Image Retrieval (CBIR) in Big Histological Image Databases
Get PDFBackground: Automatic analysis of Histopathological Images (HIs) demands image processing and Computational Intelligence (CI) techniques. Both Computer-Aided Diagnosis (CAD) and Content-Based Image-Retrieval (CBIR) systems assist diagnosis, disease discovery, and biological decision-making. Classical tests comprise screening examinations and biopsy. Histopathology slides offer more ample diagnosis data. However, manual examination of microscopic images is labor-intensive and time-consuming and may depend on a subjective assessment by the pathologist, which can be a challenge.
Methods: This work discusses a CBIR framework to extract and handle histological data, histological metadata, integrated patient records, specimen metadata, attributes, and similar stored files. This work presents a scalable image-retrieval framework for intelligent HI analysis with real-time retrieval. The potential applications of this framework include image-guided diagnosis, decision support, healthcare education, and efficient biological data management.
Results: The considerable amount of biological-related data prompted the development and deployment of large-scale databases and data-driven techniques to bridge the semantic gap between images and diagnostic information. The new cloud computing technologies and the concept of cyber-physical systems have improved the CBIR architectures considerably. The proposed scalable architecture relies on CI and validates performance on several HIs acquired from microscopic tissues. Extensive assessments show improvements in terms of disease classification and retrieval tests.
Conclusion: This research effort significant contributions are twofold. 1) Defining a comprehensive and large-scale CBIR framework to analyze HIs with high-dimensional features and CI methods successfully. 2) high-performance updating and optimization strategies improve the querying while better handling new training samples than traditional methods
Familial Mediterranean Fever in Algeria - A Retrospective of Three Molecular Studies
Get PDFBackground: Familial Mediterranean fever (FMF, OMIM 249100) is an autosomal recessive disease characterized by episodic febrile attacks and polyserositis. Renal AA-amyloidosis can complicate FMF. MEFV is the gene responsible for FMF and is involved in the regulation of inflammation. Although FMF is endemic in the Mediterranean region, its diagnosis is very recent in Algeria. We present here a retrospective of three genetic studies carried out on the Algerian patients.
Methods: 183 unrelated patients with symptoms suggestive of FMF were recruited from various hospitals between 2007 and 2015. Molecular studies included three cohorts of patients: 71 (35 males, 36 females), 84 (42 males, 42 females), and 28 (15 males, 13 females) with renal AA-amyloidosis. We searched for mutations in exon 10 of the MEFV gene by allele-specific PCR (p.M694V, p.M694I, p.M680I and p.A744S) and by resequencing the entire coding region of the same exon after PCR amplification.
Results: Molecular analysis identified 152 mutant alleles among 94/183 patients (51.36%). p.M694I was the most predominant mutation accounting for 63.2% of mutated alleles, followed by p.M694V (15.13%), p.M680I (13.81%), p.I692del (1.32%) and p.K695R (0.66%). More importantly, the M694I/M694I genotype was the most prevalent among the patients with AA-amyloidosis than the M694V/M694V genotype.
Conclusion: Our results provide the first genetic data concerning FMF in Algeria. They show that p.M694I mutation could be responsible for the severe phenotype for Algerian FMF patients
NOD2 Gene Status in Pediatric and Adult Crohn Disease Patients in Algerian Population
Get PDFBackground: Chronic Inflammatory Bowel Diseases (IBD), including Crohn disease (CD) and ulcerative colitis (UC) are gastrointestinal disorders under the influence of a complex genetic basis. One hundred sixty-three predisposition loci were identified by genome-wide association (GWAS) studies, refocusing the pathogenesis of IBD on immunity genes. The NOD2 gene has been widely implicated in the pathogenesis of IBD in different geographical populations. Three most common mutations within NOD2 gene were selected, namely SNP8, C/T (R702W variant), SNP12, G/C (G908R variant) and SNP13, (1007fsinsC variant). We investigated these three SNP in a pediatric Algerian cohort for the first time, since no previous association studies between pediatric IBD and the NOD2 gene were available for the Algerian population.
Methods: A case-control study was performed in the pediatric IBD population. PCR-RFLP was used to detect the three NOD2 gene mutations in 46 CD patients and 100 healthy control subjects. All samples were genotyped for the NOD2 gene Polymorphisms by the PCR-RFLP method. Statistical study was performed by the Fisher exact test or Chi-2 using the GraphPad Prism 7.0 software. Then data from the pediatric cohort were compared to our precedent published data from a case-control study performed on a cohort including 132 IBD patients and 114 healthy control subjects.
Results: In the pediatric cohort, there is no statistically differences in allelic frequencies between cases and controls respectively R702W (6.36% vs. 6.38%; p=1), G908R (2.72% vs. 1.06%; p=0.6) and 1007fsinsC mutation was found neither in the CD patients nor in control. In the adult cohort, the R702W allelic variant showed the highest frequency in CD patients (8%) (p = 0.09, OR = 3.67, 95%CI: 0.48-4.87) but its frequency was also high in controls (5%) (p = 0.4; OR = 1.4; 95%CI: 0.65-3.31). Likewise, G908R and 1007fsinsC mutations showed similar frequency in CD patients and in controls (3% vs. 2%; p= 0.5; OR=1.67; 95%CI: 0.44-6.34; 2% vs.1%; p=0.4, OR=2.69; 95%CI: 0.48-14.87, respectively). The total frequency of the mutated NOD2 chromosomes was higher in adult CD patients (13%) than in pediatric CD patients (9%). In our precedent study on the adult cohort, we have confirmed that the NOD2 gene is significantly associated with a specific clinical sub-phenotype in CD, indicating that the NOD2 gene is involved in IBD susceptibility across Algerian adult population. However, we failed to show any association between the three variants of the NOD2 gene across Algerian pediatric CD patients.
Conclusion: In our precedent study, we have confirmed that the NOD2 gene is significantly associated with a specific clinical sub-phenotype in adult CD patients. Here, our results show no association of NOD2 gene variants with pediatric MC. The low penetrance of the at-risk genotypes we observed indicates that the NOD2 gene does not delineate a subgroup of simple Mendelian diseases
Prediction of senior year medical students who do not pass the graduation exam by logistic analysis using data on gender, experience of repetition, and results of previous exams
Get PDFBackground. The number of students who must repeat an academic year due to an inability to attain enough credits has been increasing in Japan. It is important for universities to be able to identify these students in advance to ensure that they pass their examinations without need of repetition. In this study, we tried to predict the likelihood of students’ repetition of their senior year using the factors of gender, experience of repetition up to the junior year, and scores on tests conducted four times before their graduation exam in the senior year. Methods.Seventy-three students belonging to the senior class of a medical technology college in Tokyo were studied. The students were divided into three groups: Group 1, composed of students who passed the graduation exam on the first attempt (n=35); Group 2, composed of students who failed to pass the graduation exam at the first attempt, but passed the graduation re-exam (n=26); and Group 3, composed of students who did not pass the graduation exam or the re-exam (n=12). Results.We found that gender was not a factor of senior-year repetition. Students who had experienced of repetition prior to junior year tended to be six times more likely to fail the graduation exam than those who did not (OR=6.52, 95% CI: 1.17 – 32.44, P=0.03). Low scores on Test 4, administered two months before the graduation exam, were associated with students who fail to pass the graduation exam (OR=15.2, 95% CI: 3.29 – 70.14, P=0.00). The graduation exam score was associated with students who fail to pass the re-exam (OR=55.2, 95% CI: 1.13 – 2679.86, P=0.04). Conclusion.This study suggests that we need to support senior-year students based on the results of pre-graduation testing, and we need to increase support for students with repetition experience before junior year
Breast cancer classification using machine learning techniques: a comparative study
Get PDFBackground: The second leading deadliest disease affecting women worldwide, after lung cancer, is breast cancer. Traditional approaches for breast cancer diagnosis suffer from time consumption and some human errors in classification. To deal with this problems, many research works based on machine learning techniques are proposed. These approaches show their effectiveness in data classification in many fields, especially in healthcare.
Methods: In this cross sectional study, we conducted a practical comparison between the most used machine learning algorithms in the literature. We applied kernel and linear support vector machines, random forest, decision tree, multi-layer perceptron, logistic regression, and k-nearest neighbors for breast cancer tumors classification. The used dataset is Wisconsin diagnosis Breast Cancer.
Results: After comparing the machine learning algorithms efficiency, we noticed that multilayer perceptron and logistic regression gave the best results with an accuracy of 98% for breast cancer classification.
Conclusion: Machine learning approaches are extensively used in medical prediction and decision support systems. This study showed that multilayer perceptron and logistic regression algorithms are performant ( good accuracy specificity and sensitivity) compared to the other evaluated algorithms
Removal of Drugs from Hospitals Wastewater by Photodegradation
Get PDFBackground: Hospital wastewater (WWs) represents a particular type of effluent, compared with urban wastewater. Hospitals generate on average 750 L of wastewater by the bed and by day so they are 2-5 times higher than urban flow rates, which refer to one inhabitant equivalent (typically included in the interval 120-250 L). This significant quantity of water per day for different purposes and services depending on the activities which take place within the structure.
Methods: In our work, we proposed tow methods; the first one is the complexation with CD for the treatment of drugs from Hospital WWs, which makes solid precipitate that can be filtered later. Our second method is the photodegradation of most drugs found in Hospital WWs with a know UV lamp and time for each type of drug elimination.
Results: Our result for the first method is the formation of a CD-drug complex as a solid precipitate which will be filtered later and eliminate the solution to be purified. For the second method, the use of the UV lamp at 300 nm gave a degradation of 70% of the drug of the solution after 30 minutes of irradiation. The follow-up of the photodegradation was carried out by UV spectroscopy.
Conclusion: The elimination of drugs from hospital waters is essential, the method we have developed on a small scale can be standardized in Algerian hospitals that do not contain hospital water treatment policies, before reaching urban waters. This congress will be the occasion to discuss this national problem and a call of installation of ministry at the level of each hospital