Fully automated system
Dotaz
Zobrazit nápovědu
The process of manual species identification is a daunting task, so much so that the number of taxonomists is seen to be declining. In order to assist taxonomists, many methods and algorithms have been proposed to develop semi-automated and fully automated systems for species identification. While semi-automated tools would require manual intervention by a domain expert, fully automated tools are assumed to be not as reliable as manual or semiautomated identification tools. Hence, in this study we investigate the accuracy of fully automated and semi-automated models for species identification. We have built fully automated and semi-automated species classification models using the monogenean species image dataset. With respect to monogeneans' morphology, they are differentiated based on the morphological characteristics of haptoral bars, anchors, marginal hooks and reproductive organs (male and female copulatory organs). Landmarks (in the semi-automated model) and shape morphometric features (in the fully automated model) were extracted from four monogenean species images, which were then classified using k-nearest neighbour and artificial neural network. In semi-automated models, a classification accuracy of 96.67 % was obtained using the k-nearest neighbour and 97.5 % using the artificial neural network, whereas in fully automated models, a classification accuracy of 90 % was obtained using the k-nearest neighbour and 98.8 % using the artificial neural network. As for the crossvalidation, semi-automated models performed at 91.2 %, whereas fully automated models performed slightly higher at 93.75 %.
High-throughput root phenotyping in the soil became an indispensable quantitative tool for the assessment of effects of climatic factors and molecular perturbation on plant root morphology, development and function. To efficiently analyse a large amount of structurally complex soil-root images advanced methods for automated image segmentation are required. Due to often unavoidable overlap between the intensity of fore- and background regions simple thresholding methods are, generally, not suitable for the segmentation of root regions. Higher-level cognitive models such as convolutional neural networks (CNN) provide capabilities for segmenting roots from heterogeneous and noisy background structures, however, they require a representative set of manually segmented (ground truth) images. Here, we present a GUI-based tool for fully automated quantitative analysis of root images using a pre-trained CNN model, which relies on an extension of the U-Net architecture. The developed CNN framework was designed to efficiently segment root structures of different size, shape and optical contrast using low budget hardware systems. The CNN model was trained on a set of 6465 masks derived from 182 manually segmented near-infrared (NIR) maize root images. Our experimental results show that the proposed approach achieves a Dice coefficient of 0.87 and outperforms existing tools (e.g., SegRoot) with Dice coefficient of 0.67 by application not only to NIR but also to other imaging modalities and plant species such as barley and arabidopsis soil-root images from LED-rhizotron and UV imaging systems, respectively. In summary, the developed software framework enables users to efficiently analyse soil-root images in an automated manner (i.e. without manual interaction with data and/or parameter tuning) providing quantitative plant scientists with a powerful analytical tool.
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
This article reports for the first time a programmable-flow-based mesofluidic platform that accommodates electric-field-driven liquid phase microextraction (μ-EME) in a fully automated mode. The miniaturized system is composed of a computer-controlled microsyringe pump and a multiposition rotary valve for handling aqueous and organic solutions at a low microliter volume and acts as a front-end to online liquid chromatographic separation. The organic membrane is automatically renewed and disposed of in every analytical cycle, thus minimizing analyte carry-over effects while avoiding analyst intervention. The proof-of-concept applicability of the automated mesofluidic device is demonstrated by the liquid chromatographic determination of nonsteriodal anti-inflammatory drugs in μ-EME processed complex samples (such as urine and influent wastewater) using online heart-cut approaches. Using 5 μL of 1-octanol, 7.5 μL of untreated sample and 7.5 μL of acceptor solution (25 mM NaOH), and 250 V for only 10 min in a stopped-flow mode, the extraction recoveries for the μ-EME of ibuprofen, ketoprofen, naproxen, and diclofenac exceed 40% in real samples. The flow-through system features moderately selective extraction regardless of the sample matrix constituents with repeatability values better than 13%.
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
This study aims to develop a fully automated imaging protocol independent system for pituitary adenoma segmentation from magnetic resonance imaging (MRI) scans that can work without user interaction and evaluate its accuracy and utility for clinical applications. We trained two independent artificial neural networks on MRI scans of 394 patients. The scans were acquired according to various imaging protocols over the course of 11 years on 1.5T and 3T MRI systems. The segmentation model assigned a class label to each input pixel (pituitary adenoma, internal carotid artery, normal pituitary gland, background). The slice segmentation model classified slices as clinically relevant (structures of interest in slice) or irrelevant (anterior or posterior to sella turcica). We used MRI data of another 99 patients to evaluate the performance of the model during training. We validated the model on a prospective cohort of 28 patients, Dice coefficients of 0.910, 0.719, and 0.240 for tumour, internal carotid artery, and normal gland labels, respectively, were achieved. The slice selection model achieved 82.5% accuracy, 88.7% sensitivity, 76.7% specificity, and an AUC of 0.904. A human expert rated 71.4% of the segmentation results as accurate, 21.4% as slightly inaccurate, and 7.1% as coarsely inaccurate. Our model achieved good results comparable with recent works of other authors on the largest dataset to date and generalized well for various imaging protocols. We discussed future clinical applications, and their considerations. Models and frameworks for clinical use have yet to be developed and evaluated.
- Klíčová slova
- Image segmentation, Machine learning, Magnetic resonance imaging, Pituitary adenoma,
- MeSH
- adenom * diagnostické zobrazování chirurgie MeSH
- lidé MeSH
- magnetická rezonanční tomografie MeSH
- nádory hypofýzy * diagnostické zobrazování chirurgie MeSH
- neuronové sítě MeSH
- počítačové zpracování obrazu metody MeSH
- prospektivní studie MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
A novel flow-programming setup based on the sequential injection principle is herein proposed for on-line monitoring of temporal events in cell permeation studies. The permeation unit consists of a Franz cell with its basolateral compartment mixed under mechanical agitation and thermostated at 37 °C. The apical compartment is replaced by commercially available Transwell inserts with a precultivated cell monolayer. The transport of drug substances across epithelial cells genetically modified with the P-glycoprotein membrane transporter (MDCKII-MDR1) is monitored on-line using rhodamine 123 as a fluorescent marker. The permeation kinetics of the marker is obtained in a fully automated mode by sampling minute volumes of solution from the basolateral compartment in short intervals (10 min) up to 4 h. The effect of a P-glycoprotein transporter inhibitor, verapamil as a model drug, on the efficiency of the marker transport across the cell monolayer is thoroughly investigated. The analytical features of the proposed flow method for cell permeation studies in real time are critically compared against conventional batch-wise procedures and microfluidic devices.
- Klíčová slova
- Fully automated system, P-glycoprotein transporter, Permeation study, Rhodamine 123, Sequential injection analysis,
- MeSH
- automatizace metody MeSH
- biologický transport MeSH
- epitelové buňky chemie metabolismus MeSH
- kinetika MeSH
- lidé MeSH
- P-glykoprotein metabolismus MeSH
- průtoková injekční analýza přístrojové vybavení metody MeSH
- rhodamin 123 chemie metabolismus MeSH
- verapamil chemie metabolismus MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
- hodnotící studie MeSH
- práce podpořená grantem MeSH
- Názvy látek
- P-glykoprotein MeSH
- rhodamin 123 MeSH
- verapamil MeSH
In this paper, we present the ImmunoDisk, a fully automated sample-to-answer centrifugal microfluidic cartridge, integrating a heterogeneous, wash-free, magnetic- and fluorescent bead-based immunoassay (bound-free phase detection immunoassay/BFPD-IA). The BFPD-IA allows the implementation of a simple fluidic structure, where the assay incubation, bead separation and detection are performed in the same chamber. The system was characterized using a C-reactive protein (CRP) competitive immunoassay. A parametric investigation on air drying of protein-coupled beads for pre-storage at room temperature is presented. The key parameters were buffer composition, drying temperature and duration. A protocol for drying two different types of protein-coupled beads with the same temperature and duration using different drying buffers is presented. The sample-to-answer workflow was demonstrated measuring CRP in 5 µL of human serum, without prior dilution, utilizing only one incubation step, in 20 min turnaround time, in the clinically relevant concentration range of 15-115 mg/L. A reproducibility assessment over three disk batches revealed an average signal coefficient of variation (CV) of 5.8 ± 1.3%. A CRP certified reference material was used for method verification with a concentration CV of 8.6%. Our results encourage future testing of the CRP-ImmunoDisk in clinical studies and its point-of-care implementation in many diagnostic applications.
- Klíčová slova
- bound-free phase, centrifugal microfluidics, immunoassay, inflammation, micro/nanoparticles, point-of-care, reagent storage,
- MeSH
- C-reaktivní protein * MeSH
- imunoanalýza metody MeSH
- indikátory a reagencie MeSH
- lidé MeSH
- mikrofluidika * MeSH
- reprodukovatelnost výsledků MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
- Názvy látek
- C-reaktivní protein * MeSH
- indikátory a reagencie MeSH
Electrochemical detection systems that provide either quantitative or sample-to-answer information are promising for various analytical applications in the emerging field of point-of-care testing (POCT). Nevertheless, in mobile POC systems optical detection is currently more preferred compared to electrochemical detection due to the insufficient robustness of electrochemical detection approaches toward "real world" use. Over the last couple of decades, screen-printed electrodes (SPEs) have emerged as a simple and low-cost electrochemical detection platform. Here, we report, firstly and solely, a novel benchtop system for the processing of electrochemical methods on SPE platforms. Our solution prevents operator errors from occurring while processing and testing SPEs, achieves an automatic processing of more than 300 electrodes per day and enables comparative testing due to the presence of two simultaneous working channels; furthermore, the SPEs used can be stored in specially-designed cartridges. This novel device helps to overcome the major disadvantages in processing SPE technology, such as a low level of automation and issues with process repeatability, making this technology more efficient and enabling faster growth in industry.
- MeSH
- elektrochemické techniky * metody MeSH
- elektrody MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
One of the biggest challenges of training deep neural network is the need for massive data annotation. To train the neural network for object detection, millions of annotated training images are required. However, currently, there are no large-scale thermal image datasets that could be used to train the state of the art neural networks, while voluminous RGB image datasets are available. This paper presents a method that allows to create hundreds of thousands of annotated thermal images using the RGB pre-trained object detector. A dataset created in this way can be used to train object detectors with improved performance. The main gain of this work is the novel method for fully automatic thermal image labeling. The proposed system uses the RGB camera, thermal camera, 3D LiDAR, and the pre-trained neural network that detects objects in the RGB domain. Using this setup, it is possible to run the fully automated process that annotates the thermal images and creates the automatically annotated thermal training dataset. As the result, we created a dataset containing hundreds of thousands of annotated objects. This approach allows to train deep learning models with similar performance as the common human-annotation-based methods do. This paper also proposes several improvements to fine-tune the results with minimal human intervention. Finally, the evaluation of the proposed solution shows that the method gives significantly better results than training the neural network with standard small-scale hand-annotated thermal image datasets.
- Klíčová slova
- IR, RGB, YOLO, data annotation, deep convolutional neural networks, object detector, thermal, transfer learning,
- Publikační typ
- časopisecké články MeSH
The development of new drug delivery platforms including the use of nanotechnology has been found of great interest in recent years. Two different loading approaches of the model antimycotic drug clotrimazole into the nanofibrous polycaprolactone and polydioxanone structures including electrospinning of a drug-polymer blend and impregnation of nanofibers with drug have been tested. The final amount of clotrimazole in the nanofibrous materials was determined by HPLC analysis and Raman spectroscopy. The electrospinning of blend approach allowed the adsorption of clotrimazole in a quantity of up to 30 % using mixtures with polymer/clotrimazole ratios from 2:1 to 8:1 (w/w). Ethanolic clotrimazole solutions with concentrations from 2.5 to 3.5 mg L-1 were used for adsorbing clotrimazole in blank nanofibers for 1-3 h with final clotrimazole content ranging from 3.0 to 5.7 %. Furthermore, a comparative liberation study including comparison with commercially available creams was carried out in low pressure flow system. The results obtained confirmed well controlled release of clotrimazole from both types of nanofibers. Compared to commercial pharmaceutical formulations containing 1 % clotrimazole where first-order release kinetics was observed, nanofibrous materials provided linear controlled release (zero-order kinetics) in the tested 3 h period.
- Klíčová slova
- Automation, Drug release, Electrospinning, Nanofibers, Polycaprolactone, Polydioxanone,
- MeSH
- klotrimazol * chemie MeSH
- léky s prodlouženým účinkem MeSH
- nanovlákna * chemie MeSH
- polymery chemie MeSH
- uvolňování léčiv MeSH
- Publikační typ
- časopisecké články MeSH
- Názvy látek
- klotrimazol * MeSH
- léky s prodlouženým účinkem MeSH
- polymery MeSH
The digital polymerase chain reaction (dPCR) is an irreplaceable variant of PCR techniques due to its capacity for absolute quantification and detection of rare deoxyribonucleic acid (DNA) sequences in clinical samples. Image processing methods, including micro-chamber positioning and fluorescence analysis, determine the reliability of the dPCR results. However, typical methods demand high requirements for the chip structure, chip filling, and light intensity uniformity. This research developed an image-to-answer algorithm with single fluorescence image capture and known image-related error removal. We applied the Hough transform to identify partitions in the images of dPCR chips, the 2D Fourier transform to rotate the image, and the 3D projection transformation to locate and correct the positions of all partitions. We then calculated each partition's average fluorescence amplitudes and generated a 3D fluorescence intensity distribution map of the image. We subsequently corrected the fluorescence non-uniformity between partitions based on the map and achieved statistical results of partition fluorescence intensities. We validated the proposed algorithms using different contents of the target DNA. The proposed algorithm is independent of the dPCR chip structure damage and light intensity non-uniformity. It also provides a reliable alternative to analyze the results of chip-based dPCR systems.
- MeSH
- algoritmy MeSH
- DNA * genetika MeSH
- počítačové zpracování obrazu * MeSH
- polymerázová řetězová reakce MeSH
- reprodukovatelnost výsledků MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
- Názvy látek
- DNA * MeSH
