|
|
Detection and localization of microbial colonies by means of deep learning algorithms
Čičatka, Michal ; Vičar, Tomáš (referee) ; Mézl, Martin (advisor)
Due to massive expansion of the mass spectrometry and constant price growth of the human labour the optimalisation of the microbial samples preparation comes into question. This master thesis deals with design and implementation of a machine learning algorithm for segmentation of images of microbial colonies cultivated on Petri dishes. This algorithm is going to be a part of a controlling software of a MBT Pathfinder device developed by the company Bruker s. r. o. that automates the process of smearing microbial colonies onto a MALDI target plates. In terms of this thesis a several models of neural networks based on the UNet, UNet++ and ENet architecture were implemented. Based on a number of experiments investigating various configurations of the networks and pre-processing of the training datatset there was chosen an ENet model with quadruplet filter count and additional convolutional block of the encoder trained on a dataset pre-processed with round mask.
|
|
|
Preparation of affinity surfaces for clinical diagnostics
Kalaninová, Zuzana ; Pompach, Petr (advisor) ; Žáček, Petr (referee)
Botulinum toxins are group of protein venoms produced by Clostridium bacteria. These toxins belong to the most toxic substances causing paralytic disease called botulism. The endopeptidase activity of the toxins leads to the blockage of neurotransmitters release to neural junctions. Since the first symptoms of botulism are observed within 48 hours and toxin remains active for months, botulinum toxins are considered as potential bio- terrorism agents. For successful treatment, fast assays for detection of the toxin presence and type are highly required. This work presents a new method that could be potentially used for determination of botulotoxin in various matrices. The method is based on the detection of products of botulinum toxin A (BoNT/A) endopeptidase activity by affinity surfaces prepared by ambient ion soft landing and MALDI mass spectrometry. For immobilization and subsequent enrichment of biotinylated products of BoNT/A activity, proteins with high affinity to biotin - avidin, neutravidin and streptavidin - were used. The presented method, which combines affinity surfaces with MALDI mass spectrometry is sensitive, robust and fast and might be potentially used for detection of BoNT/A activity. Key words: ambient ion soft landing, neutravidin, botulinum toxin A, BoNT/A, MALDI/TOF mass...
|
|
|
Quantification of procalcitonin in human serum by affinity surfaces and mass spectrometry
Dvořák, Josef ; Pompach, Petr (advisor) ; Kozlík, Petr (referee)
Sepsis is a worldwide health condition. It is caused by disproportionatly large immunity system response for patogene presence. Fast and accurate diagnosis of sepsis can make difference between life and death. For clinical diagnosis of sepsis is used serum protein procalcitonin (PCT) as biomarker. Procalcitonin concentration level in patient bloodstream increases up to thousand times in short time period. Severity of sepsis can be determined by correlation with its concentration in bloodstream. To determine PCT level in patient bloodstream a broad variety of specialized instrument is used. All methods have same principle of measurement - PCT-antibody interaction. Procalcitonin level is then quantified by calibration curve method. During these measurements non-specific protein-antibody interactions can occur and distort the quantification of PCT level. Mass spectrometry due to its properties comes into consideration as an alternative method, that can be used for PCT determination. Aim of this diploma thesis was in situ enrichment of PCT on surface modified affinity carriers with immobilized antibody against PCT. These carriers are compatible with matrix assisted laser desorption/ionization time of flight mass spectrometry. Mass spectrometry provides mass spectrum, where PCT and other signals...
|
|
|
Detection and localization of microbial colonies by means of deep learning algorithms
Čičatka, Michal ; Vičar, Tomáš (referee) ; Mézl, Martin (advisor)
Due to massive expansion of the mass spectrometry and constant price growth of the human labour the optimalisation of the microbial samples preparation comes into question. This master thesis deals with design and implementation of a machine learning algorithm for segmentation of images of microbial colonies cultivated on Petri dishes. This algorithm is going to be a part of a controlling software of a MBT Pathfinder device developed by the company Bruker s. r. o. that automates the process of smearing microbial colonies onto a MALDI target plates. In terms of this thesis a several models of neural networks based on the UNet, UNet++ and ENet architecture were implemented. Based on a number of experiments investigating various configurations of the networks and pre-processing of the training datatset there was chosen an ENet model with quadruplet filter count and additional convolutional block of the encoder trained on a dataset pre-processed with round mask.
|
|
|
Quantification of procalcitonin in human serum by affinity surfaces and mass spectrometry
Dvořák, Josef ; Pompach, Petr (advisor) ; Kozlík, Petr (referee)
Sepsis is a worldwide health condition. It is caused by disproportionatly large immunity system response for patogene presence. Fast and accurate diagnosis of sepsis can make difference between life and death. For clinical diagnosis of sepsis is used serum protein procalcitonin (PCT) as biomarker. Procalcitonin concentration level in patient bloodstream increases up to thousand times in short time period. Severity of sepsis can be determined by correlation with its concentration in bloodstream. To determine PCT level in patient bloodstream a broad variety of specialized instrument is used. All methods have same principle of measurement - PCT-antibody interaction. Procalcitonin level is then quantified by calibration curve method. During these measurements non-specific protein-antibody interactions can occur and distort the quantification of PCT level. Mass spectrometry due to its properties comes into consideration as an alternative method, that can be used for PCT determination. Aim of this diploma thesis was in situ enrichment of PCT on surface modified affinity carriers with immobilized antibody against PCT. These carriers are compatible with matrix assisted laser desorption/ionization time of flight mass spectrometry. Mass spectrometry provides mass spectrum, where PCT and other signals...
|
|
|
Preparation of affinity surfaces for clinical diagnostics
Kalaninová, Zuzana ; Pompach, Petr (advisor) ; Žáček, Petr (referee)
Botulinum toxins are group of protein venoms produced by Clostridium bacteria. These toxins belong to the most toxic substances causing paralytic disease called botulism. The endopeptidase activity of the toxins leads to the blockage of neurotransmitters release to neural junctions. Since the first symptoms of botulism are observed within 48 hours and toxin remains active for months, botulinum toxins are considered as potential bio- terrorism agents. For successful treatment, fast assays for detection of the toxin presence and type are highly required. This work presents a new method that could be potentially used for determination of botulotoxin in various matrices. The method is based on the detection of products of botulinum toxin A (BoNT/A) endopeptidase activity by affinity surfaces prepared by ambient ion soft landing and MALDI mass spectrometry. For immobilization and subsequent enrichment of biotinylated products of BoNT/A activity, proteins with high affinity to biotin - avidin, neutravidin and streptavidin - were used. The presented method, which combines affinity surfaces with MALDI mass spectrometry is sensitive, robust and fast and might be potentially used for detection of BoNT/A activity. Key words: ambient ion soft landing, neutravidin, botulinum toxin A, BoNT/A, MALDI/TOF mass...
|
|
|
Mass Spectrometry Imaging of Metformin in Tissue Sections
Strnad, Štěpán ; Sýkora, D. ; Vrkoslav, Vladimír ; Cvačka, Josef ; Maletínská, Lenka ; Pirník, Zdenko
Mass spectrometry imaging is a powerful technique suitable for visualization of the distribution of a wide variety of compounds within tissue sections. The main aim of the study was the development and optimization of a sample preparation procedure allowing determination of the distribution of orally dosed metformin in mice kidney sections. Metformin is the first-line medication for the treatment of type 2 diabetes. The optimization of the sample preparation step before imaging experiments included the selection of a suitable matrix and the optimization of various parameters of MALDI analysis. 2,5-dihydroxybenzoic acid was identified as the best matrix providing highest sensitivity. A sublimation method was successfully used for the matrix deposition. The highest relative concentration of metformin was found in the inner zone of kidney 30 minutes after the drug administration.
|
|
| |
|
| |
guest ::
