ORCID Profile
0000-0001-6596-436X
Current Organisations
Queensland University of Technology
,
CSIRO
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Springer Netherlands
Date: 2015
Publisher: IEEE
Date: 12-2019
Publisher: MDPI AG
Date: 08-10-2019
DOI: 10.3390/RS11192331
Abstract: Remote sensing technology in recent years has been regarded the most important source to provide substantial information for delineating the flooding extent to the disaster management authority. There have been numerous studies proposing mathematical or statistical classification models for flood mapping. However, conventional pixel-wise classifications methods rely on the exact match of the spectral signature to label the target pixel. In this study, we propose a fully convolutional neural networks (F-CNNs) classification model to map the flooding extent from Landsat satellite images. We utilised the spatial information from the neighbouring area of target pixel in classification. A total of 64 different models were generated and trained with a variable neighbourhood size of training s les and number of learnable filters. The training results revealed that the model trained with 3 × 3 neighbourhood sized training s les and with 32 convolutional filters achieved the best performance out of the experiments. A new set of different Landsat images covering flooded areas across Australia were used to evaluate the classification performance of the model. A comparison of our proposed classification model to the conventional support vector machines (SVM) classification model shows that the F-CNNs model was able to detect flooded areas more efficiently than the SVM classification model. For ex le, the F-CNNs model achieved a maximum precision rate (true positives) of 76.7% compared to 45.27% for SVM classification.
Publisher: IEEE
Date: 2009
Publisher: Oxford University Press (OUP)
Date: 15-04-2008
DOI: 10.1086/529523
Abstract: Background and methodsThe SMART study compared 2 strategies for using antiretroviral therapy—drug conservation (DC) and viral suppression (VS)—in 5472 human immunodeficiency virus (HIV)–infected patients with CD4+ cell counts & cells/μL. Rates and predictors of opportunistic disease or death (OD/death) and the relative risk (RR) in DC versus VS groups according to the latest CD4+ cell count and HIV RNA level are reported ResultsDuring a mean of 16 months of follow-up, DC patients spent more time with a latest CD4+ cell count & cells/μL (for DC vs. VS, 31% vs. 8%) and with a latest HIV RNA level & copies/mL (71% vs. 28%) and had a higher rate of OD/death (3.4 vs. 1.3/100 person-years) than VS patients. For periods of follow-up with a CD4+ cell count & cells/μL, rates of OD/death were increased but similar in the 2 groups (5.7 vs. 4.6/100 person-years), whereas the rates were higher in DC versus VS patients (2.3 vs. 1.0/100 person-years RR, 2.3 [95% confidence interval, 1.5–3.4]) for periods with the latest CD4+ cell count ⩾350 cells/μL—an increase explained by the higher HIV RNA levels in the DC group ConclusionsThe higher risk of OD/death in DC patients was associated with (1) spending more follow-up time with relative immunodeficiency and (2) living longer with uncontrolled HIV replication even at higher CD4+ cell counts. Ongoing HIV replication at a given CD4+ cell count places patients at an excess risk of OD/death Trial registrationClinicalTrials.gov identifier: NCT00027352
Publisher: IEEE
Date: 2008
Publisher: Springer Netherlands
Date: 2015
Publisher: Oxford University Press (OUP)
Date: 15-04-2008
DOI: 10.1086/586713
Abstract: The SMART study randomized 5,472 human immunodeficiency virus (HIV)-infected patients with CD4+ cell counts >350 cells/microL to intermittent antiretroviral therapy (ART the drug conservation [DC] group) versus continuous ART (the viral suppression [VS] group). In the DC group, participants started ART when the CD4+ cell count was or= 6 months (n=228) were analyzed. The following clinical outcomes were assessed: (i) opportunistic disease (OD) or death from any cause (OD/death) (ii) OD (fatal or nonfatal) (iii) serious non-AIDS events (cardiovascular, renal, and hepatic disease plus non-AIDS-defining cancers) and non-OD deaths and (iv) the composite of outcomes (ii) and (iii). A total of 477 participants (228 in the DC group and 249 in the VS group) were followed (mean, 18 months). For outcome (iv), 21 and 6 events occurred in the DC (7 in ART-naive participants and 14 in those who had not received ART for >or= 6 months) and VS (2 in ART-naive participants and 4 in those who had not received ART for 6 months) groups, respectively. Hazard ratios for DC vs. VS by outcome category were as follows: outcome (i), 3.47 (95% confidence interval [CI], 1.26-9.56 p=.02) outcome (ii), 3.26 (95% CI, 1.04-10.25 p=.04) outcome (iii), 7.02 (95% CI, 1.57-31.38 p=.01) and outcome (iv), 4.19 (95% CI, 1.69-10.39 p=.002 ). Initiation of ART at CD4+ cell counts >350 cells/microL compared with <250 cells/microL may reduce both OD and serious non-AIDS events. These findings require validation in a large, randomized clinical trial.
Publisher: IEEE
Date: 06-2010
Location: No location found
Location: Netherlands
Location: Australia
Location: Australia
No related grants have been discovered for Chandrama Sarker.