Artificial neural network
Dotaz
Zobrazit nápovědu
- MeSH
- kognice MeSH
- lidé MeSH
- neuronové sítě MeSH
- neuroplasticita MeSH
- percepce MeSH
- teoretické modely MeSH
- Check Tag
- lidé MeSH
An extensive amount of information is currently available to clinical specialists, ranging from details of clinical symptoms to various types of biochemical data and outputs of imaging devices. Each type of data provides information that must be evaluated and assigned to a particular pathology during the diagnostic process. To streamline the diagnostic process in daily routine and avoid misdiagnosis, artificial intelligence methods (especially computer aided diagnosis and artificial neural networks) can be employed. These adaptive learning algorithms can handle diverse types of medical data and integrate them into categorized outputs. In this paper, we briefly review and discuss the philosophy, capabilities, and limitations of artificial neural networks in medical diagnosis through selected exampl
Úvod: Umělé neuronové sítě se stávají důležitou technologií při analýze dat a jejich vliv začíná prostupovat i do oblasti medicíny. Naše pracoviště se dlouhodobě věnuje experimentální chirurgii, na to navazuje náš zájem o pokrok v ostatních oblastech moderních technologií a tím i umělých neuronových sítí. V rámci aktuálního čísla chceme prozkoumat i tento aspekt technického pokroku. Hlavním cílem je kritické zhodnocení silných i slabých stránek technologie umělých neuronových sítí s ohledem na využití v klinické a experimentální chirurgii. Metody: V článku je věnována pozornost in-silico modelování a zejména pak možnostem neuronových sítí s ohledem na zpracování obrazových dat v medicíně. V textu je krátce shrnut historický vývoj hlubokého učení neuronových sítí a základní principy jejich fungování. Dále je představena taxonomie základních řešených úloh. Zmíněny jsou i možné problémy při učení i s možnostmi jejich řešení. Výsledky: Článek poukazuje na rozličné možnosti umělých neuronových sítí v biologických aplikacích. Na řadě biomedicínských aplikací umělých neuronových sítí popisuje rozdělení a princip základních úloh strojového učení a hlubokého učení – klasifikace, detekce a segmentace. Závěr: Aplikace metod umělých neuronových sítí mají v medicíně a chirurgii značný potenciál. Obcházejí potřebu zdlouhavého subjektivního nastavování parametrů znalostním inženýrem, neboť se učí přímo z dat. Při využití nevhodně vyváženého datasetu však může docházet k neočekávaným, avšak zpětně vysvětlitelným chybám. Řešení představuje vytvoření dostatečně bohatého datasetu pro učení a ověření funkce.
Introduction: Artificial neural networks are becoming an essential technology in data analysis, and their influence is starting to permeate the field of medicine. Experimental surgery has been a long-term subject of study of our lab; this is naturally reflected in our interest in other areas of modern technologies including artificial neural networks and their advancements. In the current issue, we would like to explore this aspect of technical progress. The main goal is to critically evaluate the strengths and weaknesses of artificial neural network technology concerning its use in clinical and experimental surgery. Methods: The article is focused on in-silico modeling, particularly on the potential of neural networks in terms of image data processing in medicine. The text briefly summarizes the historical development of deep learning neural networks and their basic principles. Furthermore, basic taxonomy tasks are presented. Finally, potential learning problems and possible solutions are also mentioned. Results: The article points out various possible uses of artificial neural networks in biological applications. Several biomedical applications of artificial neural networks are used to describe the division and principles of the most common tasks of machine learning and deep learning such as classification, detection, and segmentation. Conclusion: The application of artificial neural network methods in medicine and surgery offers a considerable potential; by learning directly from the data, they make it possible to avoid lengthy and subjective setting of parameters by an expert engineer. Nevertheless, the use of an unbalanced dataset can lead to unexpected, although traceable errors. The solution is to collect a dataset large enough to enable both learning and verification of proper functionality.
- MeSH
- deep learning MeSH
- lidé MeSH
- neuronové sítě * MeSH
- počítačové zpracování obrazu MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- přehledy MeSH
Examination of semen characteristics is routinely performed for fertility status investigation of the male partner of an infertile couple as well as for evaluation of the sperm donor candidate. A useful tool for preliminary assessment of semen characteristics might be an artificial neural network. Thus, the aim of the present study was to construct an artificial neural network, which could be used for predicting the result of semen analysis based on the basic questionnaire data. On the basis of eleven survey questions two models of artificial neural networks to predict semen parameters were developed. The first model aims to predict the overall performance and profile of semen. The second network was developed to predict the concentration of sperm. The network to evaluate sperm concentration proved to be the most efficient. 92.93% of the patients in the learning process were properly qualified for the group with a correct or incorrect result, while the result for the test set was 85.71%. This study suggests that an artificial neural network based on eleven survey questions might be a valuable tool for preliminary evaluation and prediction of the semen profile.
- MeSH
- analýza spermatu * metody přístrojové vybavení MeSH
- lidé MeSH
- motilita spermií MeSH
- mužská infertilita MeSH
- neuronové sítě * MeSH
- počet spermií metody přístrojové vybavení MeSH
- průzkumy a dotazníky MeSH
- sperma * MeSH
- spermabanky MeSH
- spermie * abnormality růst a vývoj ultrastruktura MeSH
- Check Tag
- lidé MeSH
Využití umělé inteligence jako asistenční detekční metody v endoskopii se v uplynulých letech těší zvyšujícímu se zájmu. Algoritmy strojového učení slibují zefektivnění detekce polypů, a dokonce optickou lokalizaci nálezů, to vše s minimálním zaškolením endoskopisty. Praktickým cílem této studie je analýza CAD softwaru (computer-aided diagnosis) Carebot pro detekci kolorektálních polypů s využitím konvoluční neuronové sítě. Navržený binární klasifikátor pro detekci polypů dosahuje přesnosti až 98 %, specificity 0,99 a preciznosti 0,96. Současně je diskutována nezbytnost dostupnosti rozsáhlých klinických dat pro vývoj modelů na bázi umělé inteligence pro automatickou detekci adenomů a benigních neoplastických lézí.
The use of artificial intelligence as an assistive detection method in endoscopy has attracted increasing interest in recent years. Machine learning algorithms promise to improve the efficiency of polyp detection and even optical localization of findings, all with minimal training of the endoscopist. The practical goal of this study is to analyse the CAD software (computer-aided diagnosis) Carebot for colorectal polyp detection using a convolutional neural network. The proposed binary classifier for polyp detection achieves accuracy of up to 98%, specificity of 0.99 and precision of 0.96. At the same time, the need for the availability of large-scale clinical data for the development of artificial--intelligence-based models for the automatic detection of adenomas and benign neoplastic lesions is discussed.
- Klíčová slova
- prostorová lokalizace,
- MeSH
- diagnóza počítačová * MeSH
- lidé MeSH
- neuronové sítě MeSH
- polypy střeva * diagnóza MeSH
- umělá inteligence MeSH
- Check Tag
- lidé MeSH
Optimization of neural network topology, weights and neuron transfer functions for given data set and problem is not an easy task. In this article, we focus primarily on building optimal feed-forward neural network classifier for i.i.d. data sets. We apply meta-learning principles to the neural network structure and function optimization. We show that diversity promotion, ensembling, self-organization and induction are beneficial for the problem. We combine several different neuron types trained by various optimization algorithms to build a supervised feed-forward neural network called Group of Adaptive Models Evolution (GAME). The approach was tested on a large number of benchmark data sets. The experiments show that the combination of different optimization algorithms in the network is the best choice when the performance is averaged over several real-world problems.
This work aims to synthesize akaganeite nanoparticles (AKNPs) by using microwave and use them to adsorb Congo red dye (CR) from the aqueous solution. The AKNPs with an average particle size of about 50 nm in width and 100 nm in length could be fabricated in 20 min. The effects of pH, CR initial concentration, adsorption time, and adsorbent dosage on the adsorption process were investigated and the artificial neural network (ANN) was used to analyze the adsorption data. The various ANN structures were examined in training the data to find the optimal model. The structure with training function, TRAINLM; adaptation learning function, LARNGDM; transfer function, LOGSIG (in hidden layer) and PURELIN (in output layer); and 10 neutrons in hidden layer having the highest correlation (R2 = 0.996) and the lowest MSE (4.405) is the optimal ANN structure. The consistency between the experimental data and the data predicted by the ANN model showed that the behavior of the adsorption process of CR onto AKNPs under different conditions can be estimated by the ANN model. The adsorption kinetics was studied by fitting the data into pseudo-first-order, pseudo-second-order, Elovich, and intraparticle diffusion models. The results showed that the adsorption kinetics obeyed the pseudo-second-order model and governed by several steps. The adsorption isotherms at the different temperatures were studied by fitting the data to Langmuir, Freundlich, and Temkin isotherm models. The R2 obtained from the Langmuir model was above 0.9 and the highest value in three of four temperatures, suggesting that the adsorption isotherms were the best fit to the Langmuir model and the maximum adsorption capacity was estimated to be more than 150 mg/g. Thermodynamic studies suggested that the adsorption of CR onto AKNPs was a spontaneous and endothermic process and physicochemical adsorption. The obtained results indicated the potential application of microwave-synthesize AKNPs for removing organic dyes from aqueous solutions.
A lower-extremity exoskeleton can facilitate the lower limbs' rehabilitation by providing additional structural support and strength. This article discusses the design and implementation of a functional prototype of lower extremity brace actuation and its wireless communication control system. The design provides supportive torque and increases the range of motion after complications reducing muscular strength. The control system prototype facilitates elevating a leg, gradually followed by standing and slow walking. The main control modalities are based on an Artificial Neural Network (ANN). The prototype's functionality was tested by time-angle graphs. The final prototype demonstrates the potential application of the ANN in the control system of exoskeletons for joint impairment therapy.
- MeSH
- chůze MeSH
- dolní končetina MeSH
- exoskeleton * MeSH
- neuronové sítě MeSH
- točivý moment MeSH
- Publikační typ
- časopisecké články MeSH
