| 1. |
- Tran, Dien M., et al.
(author)
-
High prevalence of colonisation with carbapenem-resistant Enterobacteriaceae among patients admitted to Vietnamese hospitals : Risk factors and burden of disease
- 2019
-
In: Journal of Infection. - : Saunders Elsevier. - 0163-4453 .- 1532-2742. ; 79:2, s. 115-122
-
Journal article (peer-reviewed)abstract
- BackgroundCarbapenem-resistant Enterobacteriaceae (CRE) is an increasing problem worldwide, but particularly problematic in low- and middle-income countries (LMIC) due to limitations of resources for surveillance of CRE and infection prevention and control (IPC).MethodsA point prevalence survey (PPS) with screening for colonisation with CRE was conducted on 2233 patients admitted to neonatal, paediatric and adult care at 12 Vietnamese hospitals located in northern, central and southern Vietnam during 2017 and 2018. CRE colonisation was determined by culturing of faecal specimens on selective agar for CRE. Risk factors for CRE colonisation were evaluated. A CRE admission and discharge screening sub-study was conducted among one of the most vulnerable patient groups; infants treated at an 80-bed Neonatal ICU from March throughout June 2017 to assess CRE acquisition, hospital-acquired infection (HAI) and treatment outcome.ResultsA total of 1165 (52%) patients were colonised with CRE, most commonly Klebsiella pneumoniae (n=805), Escherichia coli (n=682) and Enterobacter spp. (n=61). Duration of hospital stay, HAI and treatment with a carbapenem were independent risk factors for CRE colonisation. The PPS showed that the prevalence of CRE colonisation increased on average 4.2 % per day and mean CRE colonisation rates increased from 13% on the day of admission to 89% at day 15 of hospital stay. At the NICU CRE colonisation increased from 32% at admission to 87% at discharge, mortality was significantly associated (OR 5•5, P < 0•01) with CRE colonisation and HAI on admission.ConclusionThese data indicate that there is an epidemic spread of CRE in Vietnamese hospitals with rapid transmission to hospitalised patients.
|
|
| 2. |
- Thang, Truong Cong, et al.
(author)
-
Design and implementation of an e-Health system for depression detection
- 2015
-
Conference paper (peer-reviewed)abstract
- We present the design and implementation of a cost-effective e-Health system for automatic depression detection. The system is based on a client-server architecture, where clients are popular mobile devices. For practical deployment, various factors that affect the accuracy and speed of depression detection are discussed and evaluated with extensive experiments.
|
|
| 3. |
- Arunachalam, Ajay, 1985-, et al.
(author)
-
Toward Data-Model-Agnostic Autonomous Machine-Generated Data Labeling and Annotation Platform : COVID-19 Autoannotation Use Case
- 2023
-
In: IEEE transactions on engineering management. - : IEEE. - 0018-9391 .- 1558-0040. ; 70:8, s. 2695-2706
-
Journal article (peer-reviewed)abstract
- Quick, early, and precise detection is important for diagnosis to control the spread of COVID-19 infection. Artificial Intelligence (AI) technology could certainly be used as a modulating tool to ease the detection, and help with the preventive steps further. Convolutional neural networks (CNNs) have achieved state-of-the-art performance in many visual recognition tasks. Nevertheless, most of these state-of-the-art networks highly rely on the availability of a high amount of labeled data, being an essential step in supervised machine learning tasks. Conventionally, this manual, mundane, and time-consuming process of annotating images is done by humans. Learning to localize or detect COVID-19 infection masks in our specific case study typically requires the collection of CT scan data that has been labeled with bounding boxes or similar annotations, which generally is limited. A technique that could perform such learning with much less annotations, and transfer the learned proposals that are algorithm-driven to generate more synthetic annotated samples would be helpful & quite valuable. We present such a technique inspired by weakly trained mask region based convolutional neural networks (R-CNN) architecture for localization, in which the number of images with their pixel-level masks can be a small proportion of the total dataset, and then further improvise CNNs by inversely generating dense annotations on-the-go using an algorithmic-based computational approach. We focus on alleviating the bottleneck associated with deep learning models needing annotated data for training in an intuitive reverse engineering fashion through this work. Our proposed solution can certainly provide the prospect of automated labeling on-the-fly, thereby reducing much of the manual work. As a result, one can quickly train a precise COVID-19 infection detector with the leverage of autonomous frame-by-frame machine generated annotations. The model achieved mean precision accuracy (%) of 0.99, 0.931, and 0.8 for train, validation, and test set, respectively. The results demonstrate that the proposed method can be adopted in a clinical setting for assisting radiologists, and also our fully autonomous approach can be generalized to any detection/recognition tasks at ease.
|
|
| 4. |
- Beck, Dominik, et al.
(author)
-
Integrative analysis of next generation sequencing for small non-coding RNAs and transcriptional regulation in Myelodysplastic Syndromes
- 2011
-
In: BMC Medical Genomics. - : Springer Science and Business Media LLC. - 1755-8794. ; 4:19, s. 1-16
-
Journal article (peer-reviewed)abstract
- BackgroundMyelodysplastic Syndromes (MDSS) are pre-leukemic disorders with increasing incident rates worldwide, but very limited treatment options. Little is known about small regulatory RNAs and how they contribute to pathogenesis, progression and transcriptome changes in MDS.MethodsPatients' primary marrow cells were screened for short RNAs (RNA-seq) using next generation sequencing. Exon arrays from the same cells were used to profile gene expression and additional measures on 98 patients obtained. Integrative bioinformatics algorithms were proposed, and pathway and ontology analysis performed.ResultsIn low-grade MDS, observations implied extensive post-transcriptional regulation via microRNAs (miRNA) and the recently discovered Piwi interacting RNAs (piRNA). Large expression differences were found for MDS-associated and novel miRNAs, including 48 sequences matching to miRNA star (miRNA*) motifs. The detected species were predicted to regulate disease stage specific molecular functions and pathways, including apoptosis and response to DNA damage. In high-grade MDS, results suggested extensive post-translation editing via transfer RNAs (tRNAs), providing a potential link for reduced apoptosis, a hallmark for this disease stage. Bioinformatics analysis confirmed important regulatory roles for MDS linked miRNAs and TFs, and strengthened the biological significance of miRNA*. The "RNA polymerase II promoters" were identified as the tightest controlled biological function. We suggest their control by a miRNA dominated feedback loop, which might be linked to the dramatically different miRNA amounts seen between low and high-grade MDS.DiscussionThe presented results provide novel findings that build a basis of further investigations of diagnostic biomarkers, targeted therapies and studies on MDS pathogenesis.
|
|
| 5. |
- Brandl, Miriam B, et al.
(author)
-
Application of Fuzzy c-Means and Joint-Feature-Clustering to Detect Redundancies of Image-Features in Drug Combinations Studies of Breast Cancer
- 2011
-
In: AIP Conference Proceedings. - : AIP. - 0094-243X.
-
Conference paper (peer-reviewed)abstract
- The high dimensionality of image‐based dataset can be a drawback for classification accuracy. In this study, we propose the application of fuzzy c‐means clustering, cluster validity indices and the notation of a joint‐feature‐clustering matrix to find redundancies of image‐features. The introduced matrix indicates how frequently features are grouped in a mutual cluster. The resulting information can be used to find data‐derived feature prototypes with a common biological meaning, reduce data storage as well as computation times and improve the classification accuracy
|
|
| 6. |
-
Knowledge-Based Systems in Biomedicine and Computational Life Science
- 2012
-
Editorial collection (other academic/artistic)abstract
- This book presents a sample of research on knowledge-based systems in biomedicine and computational life science. The contributions include: personalized stress diagnosis system, image analysis system for breast cancer diagnosis, analysis of neuronal cell images, structure prediction of protein, relationship between two mental disorders, detection of cardiac abnormalities, holistic medicine based treatment and analysis of life-science data.
|
|
| 7. |
- Liu, Jin, et al.
(author)
-
A spatially constrained fuzzy hyper-prototype clustering algorithm
- 2012
-
In: Pattern Recognition. - : Elsevier. - 0031-3203 .- 1873-5142. ; 45:4, s. 1759-1771
-
Journal article (peer-reviewed)abstract
- We present in this paper a fuzzy clustering algorithm which can handle spatially constraint problems often encountered in pattern recognition. The proposed method is based on the notions of hyperplanes, the fuzzy c-means, and spatial constraints. By adding a spatial regularizer into the fuzzy hyperplane-based objective function, the proposed method can take into account additionally important information of inherently spatial data. Experimental results have demonstrated that the proposed algorithm achieves superior results to some other popular fuzzy clustering models, and has potential for cluster analysis in spatial domain.
|
|
| 8. |
- Liu, Jin, et al.
(author)
-
FHC : The fuzzy hyper-prototype clustering algorithm
- 2012
-
In: Journal of Knowledge-based & Intelligent Engineering Systems. - : IOS Press. - 1327-2314 .- 1875-8827. ; 16:1, s. 35-47
-
Journal article (peer-reviewed)abstract
- We propose a fuzzy hyper-prototype clustering algorithm in this paper. This approach uses hyperplanes to represent the cluster centers in the fuzzy clustering. We present the formulation of fuzzy objective function and derive an iterative numerical algorithm for minimizing the objective function. Validations and comparisons are made between the proposed fuzzy clustering algorithm and existing fuzzy clustering methods on artificially generated data as well as on real world dataset include UCI dataset and gene expression dataset, the results show that the proposed method can give better performance in the above cases.
|
|
| 9. |
- Liu, Jin, et al.
(author)
-
Fuzzy hyper-prototype clustering
- 2010
-
In: Knowledge-Based and Intelligent Information and Engineering Systems. - Berlin, Heidelberg : Springer Berlin/Heidelberg. - 9783642153860 - 9783642153877 ; , s. 379-389
-
Conference paper (other academic/artistic)abstract
- We propose a fuzzy hyper-prototype algorithm in this paper. This approach uses hyperplanes to represent the cluster centers in the fuzzy c-means algorithm. We present the formulation of a hyperplane-based fuzzy objective function and then derive an iterative numerical procedure for minimizing the clustering criterion. We tested the method with data degraded with random noise. The experimental results show that the proposed method is robust to clustering noisy linear structure.
|
|
| 10. |
- Ng, Theam Foo, et al.
(author)
-
Automated feature weighting in fuzzy declustering-based vector quantization
- 2010
-
Conference paper (peer-reviewed)abstract
- Feature weighting plays an important role in improving the performance of clustering technique. We propose an automated feature weighting in fuzzy declustering-based vector quantization (FDVQ), namely AFDVQ algorithm, for enhancing effectiveness and efficiency in classification. The proposed AFDVQ imposes weights on the modified fuzzy c-means (FCM) so that it can automatically calculate feature weights based on their degrees of importance rather than treating them equally. Moreover, the extension of FDVQ and AFDVQ algorithms based on generalized improved fuzzy partitions (GIFP), known as GIFP-FDVQ and GIFP-AFDVQ respectively, are proposed. The experimental results on real data (original and noisy data) and modified data (biased and noisy-biased data) have demonstrated that the proposed algorithms outperformed standard algorithms in classifying clusters especially for biased data.
|
|
