Algorithm comparison
Dotaz
Zobrazit nápovědu
MOTIVATION: Automatic tracking of cells in multidimensional time-lapse fluorescence microscopy is an important task in many biomedical applications. A novel framework for objective evaluation of cell tracking algorithms has been established under the auspices of the IEEE International Symposium on Biomedical Imaging 2013 Cell Tracking Challenge. In this article, we present the logistics, datasets, methods and results of the challenge and lay down the principles for future uses of this benchmark. RESULTS: The main contributions of the challenge include the creation of a comprehensive video dataset repository and the definition of objective measures for comparison and ranking of the algorithms. With this benchmark, six algorithms covering a variety of segmentation and tracking paradigms have been compared and ranked based on their performance on both synthetic and real datasets. Given the diversity of the datasets, we do not declare a single winner of the challenge. Instead, we present and discuss the results for each individual dataset separately. AVAILABILITY AND IMPLEMENTATION: The challenge Web site (http://www.codesolorzano.com/celltrackingchallenge) provides access to the training and competition datasets, along with the ground truth of the training videos. It also provides access to Windows and Linux executable files of the evaluation software and most of the algorithms that competed in the challenge.
Four methods for global numerical black box optimization with origins in the mathematical programming community are described and experimentally compared with the state of the art evolutionary method, BIPOP-CMA-ES. The methods chosen for the comparison exhibit various features that are potentially interesting for the evolutionary computation community: systematic sampling of the search space (DIRECT, MCS) possibly combined with a local search method (MCS), or a multi-start approach (NEWUOA, GLOBAL) possibly equipped with a careful selection of points to run a local optimizer from (GLOBAL). The recently proposed "comparing continuous optimizers" (COCO) methodology was adopted as the basis for the comparison. Based on the results, we draw suggestions about which algorithm should be used depending on the available budget of function evaluations, and we propose several possibilities for hybridizing evolutionary algorithms (EAs) with features of the other compared algorithms.
- MeSH
- Albers-Schönbergova nemoc MeSH
- algoritmy MeSH
- alkalická fosfatasa analýza MeSH
- biochemická analýza krve MeSH
- biologické markery analýza MeSH
- denzitometrie MeSH
- diabetes mellitus MeSH
- diagnóza počítačová * MeSH
- dospělí MeSH
- klinické laboratorní techniky MeSH
- kosti a kostní tkáň metabolismus MeSH
- lidé středního věku MeSH
- lidé MeSH
- metabolické nemoci kostí * diagnóza etiologie prevence a kontrola terapie MeSH
- osteogeneze * účinky léků MeSH
- osteokalcin analýza MeSH
- osteoporóza MeSH
- rizikové faktory MeSH
- sbírání vzorku moči MeSH
- sulfonylmočovinové sloučeniny škodlivé účinky MeSH
- výsledek terapie * MeSH
- zátěžový test MeSH
- zinek nedostatek MeSH
- Check Tag
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- mužské pohlaví MeSH
- ženské pohlaví MeSH
- Publikační typ
- přednášky MeSH
- srovnávací studie MeSH
PURPOSE: The aim of this work was to introduce a new algorithm for image reconstruction in bone SPECT and to compare its performances with a commercially available standard OSEM and resolution recovery (RR) reconstruction. MATERIALS AND METHODS: The algorithm was built applying the Lucy-Richardson deconvolution adn logarithmic image processing to the projections. A modification of the coefficients of wavelet decomposition was used to suppress the noise. The comparison with vendor software was performed both in a phantom study, using Signal-to-Noise ratio (SNR), Signal-to-Background ratio (SBR), spatial resolution and in clinical studies, by visual assessment of changes in contrast, spatial resolution and lesion detectability. RESULTS: A change in the SNR (from -4 to 40%), an increase in the SBR (from 19 to 40%), a minor improvement in spatial resolution and a similar noise level were observed in the phantom study in comparison to the standard OSEM. A decrease in the SNR, a worse spatial resolution, but only a 3 to 13 % lower SBR were achieved in comparison with the vendor supplied RR algorithm. The proposed algorithm creates patient images with better contrast and lesion detectability compared to clinically used OSEM. Compared to RR, more than half of obtained images showed better contrast and nearly half of them have better lesion detectability. CONCLUSION: The proposed algorithm compares favorably with the standard OSEM. Although less favorable, the comparison with RR and noise suppression algorithms, suggests that it can be used with only a slight decrease in the SBR.
- MeSH
- algoritmy * MeSH
- fantomy radiodiagnostické * MeSH
- jednofotonová emisní výpočetní tomografie přístrojové vybavení MeSH
- kosti a kostní tkáň MeSH
- lidé MeSH
- počítačové zpracování obrazu metody MeSH
- poměr signál - šum MeSH
- vlnková analýza * MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
In this paper, we analysed the steady state fluorescence spectra of cell suspensions containing healthy and carcinoma fibroblast mouse cells, using a genetic-algorithm-spectra-decomposition software (GASpeD). In contrast to other deconvolution algorithms, such as polynomial or linear unmixing software, GASpeD takes into account light scatter. In cell suspensions, light scatter plays an important role as it depends on the number of cells, their size, shape, and coagulation. The measured fluorescence spectra were normalized, smoothed and deconvoluted into four peaks and background. The wavelengths of intensities' maxima of lipopigments (LR), FAD, and free/bound NAD(P)H (AF/AB) of the deconvoluted spectra matched published data. In deconvoluted spectra at pH = 7, the fluorescence intensities of the AF/AB ratio in healthy cells was always higher in comparison to carcinoma cells. In addition, the AF/AB ratio in healthy and carcinoma cells were influenced differently by changes in pH. In mixtures of healthy and carcinoma cells, AF/AB decreases when more than 13% of carcinoma cells are present. Expensive instrumentation is not required, and the software is user friendly. Due to these attributes, we hope that this study will be a first step in the development of new cancer biosensors and treatments with the use of optical fibers.
- MeSH
- algoritmy * MeSH
- buněčné kultury MeSH
- fluorescence MeSH
- fluorescenční spektrometrie MeSH
- karcinom * MeSH
- myši MeSH
- software MeSH
- zvířata MeSH
- Check Tag
- myši MeSH
- zvířata MeSH
- Publikační typ
- časopisecké články MeSH
- MeSH
- algoritmy MeSH
- finanční podpora výzkumu jako téma MeSH
- hyperglykemie terapie MeSH
- inzulin aplikace a dávkování MeSH
- krevní glukóza analýza MeSH
- lidé MeSH
- monitorování fyziologických funkcí MeSH
- péče o pacienty v kritickém stavu MeSH
- pilotní projekty MeSH
- systémy podporující rozhodování v léčbě MeSH
- Check Tag
- lidé MeSH
Background: Type-1 diabetes is a condition caused by the lack of insulin hormone, which leads to an excessive increase in blood glucose level. The glucose kinetics process is difficult to control due to its complex and nonlinear nature and with state variables that are difficult to measure. Methods: This paper proposes a method for automatically calculating the basal and bolus insulin doses for patients with type-1 diabetes using reinforcement learning with feedforward controller. The algorithm is designed to keep the blood glucose stable and directly compensate for the external events such as food intake. Its performance was assessed using simulation on a blood glucose model. The usage of the Kalman filter with the controller was demonstrated to estimate unmeasurable state variables. Results: Comparison simulations between the proposed controller with the optimal reinforcement learning and the proportional-integral-derivative controller show that the proposed methodology has the best performance in regulating the fluctuation of the blood glucose. The proposed controller also improved the blood glucose responses and prevented hypoglycemia condition. Simulation of the control system in different uncertain conditions provided insights on how the inaccuracies of carbohydrate counting and meal-time reporting affect the performance of the control system. Conclusion: The proposed controller is an effective tool for reducing postmeal blood glucose rise and for countering the effects of external known events such as meal intake and maintaining blood glucose at a healthy level under uncertainties.
- MeSH
- algoritmy * MeSH
- biologické modely MeSH
- diabetes mellitus 1. typu krev farmakoterapie MeSH
- inzulin aplikace a dávkování MeSH
- kinetika MeSH
- krevní glukóza metabolismus MeSH
- lidé MeSH
- počítačem asistovaná terapie statistika a číselné údaje MeSH
- počítačová simulace MeSH
- posilování (psychologie) MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
- hodnotící studie MeSH
Proper classification of action potentials from extracellular recordings is essential for making an accurate study of neuronal behavior. Many spike sorting algorithms have been presented in the technical literature. However, no comparative analysis has hitherto been performed. In our study, three widely-used publicly-available spike sorting algorithms (WaveClus, KlustaKwik, OSort) were compared with regard to their parameter settings. The algorithms were evaluated using 112 artificial signals (publicly available online) with 2-9 different neurons and varying noise levels between 0.00 and 0.60. An optimization technique based on Adjusted Mutual Information was employed to find near-optimal parameter settings for a given artificial signal and algorithm. All three algorithms performed significantly better (p<0.01) with optimized parameters than with the default ones. WaveClus was the most accurate spike sorting algorithm, receiving the best evaluation score for 60% of all signals. OSort operated at almost five times the speed of the other algorithms. In terms of accuracy, OSort performed significantly less well (p<0.01) than WaveClus for signals with a noise level in the range 0.15-0.30. KlustaKwik achieved similar scores to WaveClus for signals with low noise level 0.00-0.15 and was worse otherwise. In conclusion, none of the three compared algorithms was optimal in general. The accuracy of the algorithms depended on proper choice of the algorithm parameters and also on specific properties of the examined signal.
- MeSH
- akční potenciály fyziologie MeSH
- algoritmy MeSH
- elektrofyziologie metody MeSH
- lidé MeSH
- neurony fyziologie MeSH
- počítačové zpracování signálu MeSH
- validace softwaru MeSH
- zvířata MeSH
- Check Tag
- lidé MeSH
- zvířata MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
- srovnávací studie MeSH
- validační studie MeSH
Cíl. Cílem studie bylo posoudit možnosti IRIS (iterativní rekonstrukce v obrazovém prostoru) algoritmu z hlediska redukce šumu v obraze a efektivní dávky. Metoda. Bylo vyhodnoceno celkem 85 pacientů, kteří podstoupili CT vyšetření (plic, břicha a mozku) s rekonstrukcí snímků pomocí IRIS algoritmu, a podobná skupina 95 pacientů, jejichž snímky byly rekonstruovány filtrovanou zpětnou projekcí (FBP). Referenční součin proudu a expozičního času byl pro skupinu s IRIS algoritmem v původních užívaných protokolech snižován tak, aby kvalita snímku na základě subjektivního posouzení radiologa byla zachována. Obě skupiny byly porovnány z hlediska množství šumu na snímku, poměru signálu a šumu a efektivní dávky, kterou pacient během vyšetření obdrží. Výsledky. Ukázalo se, že u vyšetření břicha je možné radiační zátěž pacienta snížit až o 50 %, aniž by došlo ke zhoršení kvality snímku. Na druhou stranu pro vyšetření mozku se podařilo dosáhnout snížení průměrné efektivní dávky pouze o 23 %. Závěr. Z tohoto důvodu by měla obdobná měření a optimalizace vždy předcházet uvedení nového rekonstrukčního algoritmu do klinické praxe.
Aim. The aim of this study was to assess the possibilities of IRIS (Iterative reconstruction in Image Space) algorithm in noise reduction and potential dose reduction. Method. A total of 85 patients clinically underwent CT examination (chest, abdomen, brain) with IRIS reconstruction and similar groups of 95 patients with Filtered Back Projection (FBP) reconstruction were evaluated. The reference tube current product in original CT protocols was modified for IRIS algorithm according to the subjective image quality. Two reconstruction algorithms were compared in terms of image noise, signal to noise ratio and effective dose. Results. It was demonstrated that with the novel approach the mean effective dose reduction up to 50% for CT examination of the abdomen did not lead to image quality degradation. However, to retain the quality of CT images of the brain it was accomplished mean effective dose reduction only of 23%. Conclusion. For this reason such measurements and similar optimization should precede to introduction of any new reconstruction algorithm into clinical routine.
- Klíčová slova
- CT, efektivní dávka,
- MeSH
- algoritmy MeSH
- artefakty MeSH
- dávka záření MeSH
- diagnostické zobrazování metody přístrojové vybavení trendy MeSH
- lidé MeSH
- počítačová rentgenová tomografie metody trendy MeSH
- počítačové zpracování obrazu metody trendy využití MeSH
- řízení kvality MeSH
- software trendy MeSH
- statistika jako téma MeSH
- tělesná hmotnost MeSH
- výsledky a postupy - zhodnocení (zdravotní péče) MeSH
- Check Tag
- lidé MeSH
The Convolution algorithm, implemented in Leksell GammaPlan® ver. Here, 10, is the first algorithm for Leksell Gamma Knife that takes heterogeneities into account and models dose build-up effects close to tissue boundaries. The aim of this study was preliminary comparison of the Convolution and TMR10 algorithms for real clinical cases and dosimetric verification of the algorithms, using measurements in a phantom. A total of 25 patients involved in comparison of the Convolution and TMR10 algorithms were divided into three groups: patients with benign tumors close to heterogeneities, patients with functional disorders, and patients with tumors located far from heterogeneities. Differences were observed especially in the group of patients with tumors close to heterogeneities, where the difference in maximal dose to critical structures for the Convolution algorithm was up to 15% compared to the TMR10 algorithm. Dosimetric verification of the algorithm was performed, using a radiochromic gel dosimeter based on Turnbull blue dye in a special heterogeneous phantom. Relative dose distributions measured with the radiochromic gel dosimeter agreed very well with both the TMR10 and Convolution calculations. We observed small discrepancies in the direction in which the largest inhomogeneity was positioned. Verification results indicated that the Convolution algorithm provides a different dose distribution, especially in regions close to heterogeneities and particularly for lower isodose volumes. However, the results obtained with gamma analyses in the gel dosimetry experiment did not verify the assumption that the Convolution algorithm provides more accurate dose calculation.
- MeSH
- algoritmy * MeSH
- celková dávka radioterapie MeSH
- fantomy radiodiagnostické * MeSH
- filmová dozimetrie * MeSH
- kritické orgány účinky záření MeSH
- lidé MeSH
- metoda Monte Carlo MeSH
- nádory chirurgie MeSH
- plánování radioterapie pomocí počítače metody MeSH
- radiochirurgie metody MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
- srovnávací studie MeSH
