Tram Anh Nguyen - Academia.edu (original) (raw)

Papers by Tram Anh Nguyen

Fluorescence microscopy images are often taken at low light and short exposure times to preserve ... more Fluorescence microscopy images are often taken at low light and short exposure times to preserve the integrity of cell samples. However, imaging under these conditions leads to severely degraded images with low signal to noise ratios. To computationally restore these images, we introduce novel loss functions to denoise microscopy images. These loss functions will be folded into the CARE algorithm. The results produced by this modification will be evaluated against traditional TV filtering and NL means techniques. The modified model will also be compared against its CARE predecessor using standard image quality metrics.

BackgroundFluorescence microscopy is an important technique in many areas of biological research.... more BackgroundFluorescence microscopy is an important technique in many areas of biological research. Two factors which limit the usefulness and performance of fluorescence microscopy are photobleaching of fluorescent probes during imaging, and, when imaging live cells, phototoxicity caused by light exposure. Recently developed methods in machine learning are able to greatly improve the signal to noise ratio of acquired images. This allows researchers to record images with much shorter exposure times, which in turn minimizes photobleaching and phototoxicity by reducing the dose of light reaching the sample.FindingsTo employ deep learning methods, a large amount of data is needed to train the underlying convolutional neural network. One way to do this involves use of pairs of fluorescence microscopy images acquired with long and short exposure times. We provide high quality data sets which can be used to train and evaluate deep learning methods under development.ConclusionThe availabilit...

Tropical Medicine & International Health, 2008

Method From November 2004 to October 2005, water and herb samples collected from markets and farm... more Method From November 2004 to October 2005, water and herb samples collected from markets and farms in Hanoi were examined for the presence of Cyclospora spp. oocysts in concentrated sediments and washings using UV epifluorescence examination of a wet mount. In ...

Tropical Medicine & International Health, 2007

To identify the level of contamination with thermotolerant coliforms (ThC), intestinal helminth e... more To identify the level of contamination with thermotolerant coliforms (ThC), intestinal helminth eggs and protozoan parasites in water spinach (Ipomoea aquatica) cultivated in a wastewater-fed lake in Phnom Penh, Cambodia. The investigation was carried out from July 2004 to May 2005 in Boeng Cheung Ek lake, a main recipient of wastewater from the city of Phnom Penh. Water spinach samples were collected near two wastewater entry points in the northern part of the lake and near a wastewater exit point in the southern part of the lake, where water is discharged from the lake into a stream. Water spinach samples were harvested at each of the three locations with and without wastewater contact according to the normal practices of farmers, and analysed for ThC, protozoan parasites (Giardia, Cryptosporidium and Cyclospora), and helminth eggs (Ascaris lumbricoides, Trichuris trichura and hookworm) using standard methods. Wastewater samples were also collected at each of the three locations and analysed for ThC and helminth eggs. High concentrations of ThC (approximately 10(5)-10(7)/g) were found in water spinach samples. ThC mean counts did not differ significantly, neither between sample types at each location, nor between locations. ThC mean concentration of water spinach samples collected in the rainy season of 2005 was higher than that in the dry season of 2004. Protozoan parasites were detected in the samples from all three locations and in water spinach harvested with and without wastewater contact. Giardia spp. were present in 56% (20) of samples with an average cyst concentration of 6.6/g. Mean counts of Giardia were not significantly associated with wastewater contact status of water spinach. Cryptosporidium oocysts were found in 17% (six) of samples with an average concentration of 0.5 oocyst per gram of water spinach. Cyclospora oocysts were detected in 8% (three) of samples. Four (11%) water spinach samples contained helminth eggs (average concentration: 0.1 egg per gram). Water samples contained approximately 10(4)-10(7) ThC per 100 ml and a mean of 27.4 helminth eggs per litre. There was a 2-3 log reduction of ThC between the water samples collected at the wastewater entry and exit point(s) of the lake. Water spinach cultivated in the lake was highly contaminated with feces, as indicated by high ThC concentrations and the presence of protozoan parasites. A reduction in ThC numbers almost to WHO guideline levels for irrigation water occurred through natural biological and physical processes in the lake, as shown by differences in ThC counts at wastewater entry and exit points.

Fluorescence microscopy images are often taken at low light and short exposure times to preserve ... more Fluorescence microscopy images are often taken at low light and short exposure times to preserve the integrity of cell samples. However, imaging under these conditions leads to severely degraded images with low signal to noise ratios. To computationally restore these images, we introduce novel loss functions to denoise microscopy images. These loss functions will be folded into the CARE algorithm. The results produced by this modification will be evaluated against traditional TV filtering and NL means techniques. The modified model will also be compared against its CARE predecessor using standard image quality metrics.

BackgroundFluorescence microscopy is an important technique in many areas of biological research.... more BackgroundFluorescence microscopy is an important technique in many areas of biological research. Two factors which limit the usefulness and performance of fluorescence microscopy are photobleaching of fluorescent probes during imaging, and, when imaging live cells, phototoxicity caused by light exposure. Recently developed methods in machine learning are able to greatly improve the signal to noise ratio of acquired images. This allows researchers to record images with much shorter exposure times, which in turn minimizes photobleaching and phototoxicity by reducing the dose of light reaching the sample.FindingsTo employ deep learning methods, a large amount of data is needed to train the underlying convolutional neural network. One way to do this involves use of pairs of fluorescence microscopy images acquired with long and short exposure times. We provide high quality data sets which can be used to train and evaluate deep learning methods under development.ConclusionThe availabilit...

Tropical Medicine & International Health, 2008

Method From November 2004 to October 2005, water and herb samples collected from markets and farm... more Method From November 2004 to October 2005, water and herb samples collected from markets and farms in Hanoi were examined for the presence of Cyclospora spp. oocysts in concentrated sediments and washings using UV epifluorescence examination of a wet mount. In ...

Tropical Medicine & International Health, 2007

To identify the level of contamination with thermotolerant coliforms (ThC), intestinal helminth e... more To identify the level of contamination with thermotolerant coliforms (ThC), intestinal helminth eggs and protozoan parasites in water spinach (Ipomoea aquatica) cultivated in a wastewater-fed lake in Phnom Penh, Cambodia. The investigation was carried out from July 2004 to May 2005 in Boeng Cheung Ek lake, a main recipient of wastewater from the city of Phnom Penh. Water spinach samples were collected near two wastewater entry points in the northern part of the lake and near a wastewater exit point in the southern part of the lake, where water is discharged from the lake into a stream. Water spinach samples were harvested at each of the three locations with and without wastewater contact according to the normal practices of farmers, and analysed for ThC, protozoan parasites (Giardia, Cryptosporidium and Cyclospora), and helminth eggs (Ascaris lumbricoides, Trichuris trichura and hookworm) using standard methods. Wastewater samples were also collected at each of the three locations and analysed for ThC and helminth eggs. High concentrations of ThC (approximately 10(5)-10(7)/g) were found in water spinach samples. ThC mean counts did not differ significantly, neither between sample types at each location, nor between locations. ThC mean concentration of water spinach samples collected in the rainy season of 2005 was higher than that in the dry season of 2004. Protozoan parasites were detected in the samples from all three locations and in water spinach harvested with and without wastewater contact. Giardia spp. were present in 56% (20) of samples with an average cyst concentration of 6.6/g. Mean counts of Giardia were not significantly associated with wastewater contact status of water spinach. Cryptosporidium oocysts were found in 17% (six) of samples with an average concentration of 0.5 oocyst per gram of water spinach. Cyclospora oocysts were detected in 8% (three) of samples. Four (11%) water spinach samples contained helminth eggs (average concentration: 0.1 egg per gram). Water samples contained approximately 10(4)-10(7) ThC per 100 ml and a mean of 27.4 helminth eggs per litre. There was a 2-3 log reduction of ThC between the water samples collected at the wastewater entry and exit point(s) of the lake. Water spinach cultivated in the lake was highly contaminated with feces, as indicated by high ThC concentrations and the presence of protozoan parasites. A reduction in ThC numbers almost to WHO guideline levels for irrigation water occurred through natural biological and physical processes in the lake, as shown by differences in ThC counts at wastewater entry and exit points.