Artif Intell Med - On the interplay of machine learning and background knowledge in image interpretation by Bayesian networks.

Tópicos

{ model(3404) distribut(989) bayesian(671) }
{ gene(2352) biolog(1181) express(1162) }
{ system(1976) rule(880) can(841) }
{ structur(1116) can(940) graph(676) }
{ featur(3375) classif(2383) classifi(1994) }
{ data(3008) multipl(1320) sourc(1022) }
{ monitor(1329) mobil(1314) devic(1160) }
{ cancer(2502) breast(956) screen(824) }
{ learn(2355) train(1041) set(1003) }
{ perform(1367) use(1326) method(1137) }
{ data(1737) use(1416) pattern(1282) }
{ method(1219) similar(1157) match(930) }
{ patient(2315) diseas(1263) diabet(1191) }
{ surgeri(1148) surgic(1085) robot(1054) }
{ chang(1828) time(1643) increas(1301) }
{ control(1307) perform(991) simul(935) }
{ howev(809) still(633) remain(590) }
{ import(1318) role(1303) understand(862) }
{ first(2504) two(1366) second(1323) }
{ can(981) present(881) function(850) }
{ imag(2830) propos(1344) filter(1198) }
{ imag(2675) segment(2577) method(1081) }
{ data(3963) clinic(1234) research(1004) }
{ perform(999) metric(946) measur(919) }
{ method(2212) result(1239) propos(1039) }
{ can(774) often(719) complex(702) }
{ imag(1057) registr(996) error(939) }
{ sequenc(1873) structur(1644) protein(1328) }
{ problem(2511) optim(1539) algorithm(950) }
{ error(1145) method(1030) estim(1020) }
{ design(1359) user(1324) use(1319) }
{ model(2220) cell(1177) simul(1124) }
{ studi(1410) differ(1259) use(1210) }
{ blood(1257) pressur(1144) flow(957) }
{ time(1939) patient(1703) rate(768) }
{ result(1111) use(1088) new(759) }
{ survey(1388) particip(1329) question(1065) }
{ activ(1452) weight(1219) physic(1104) }
{ imag(1947) propos(1133) code(1026) }
{ inform(2794) health(2639) internet(1427) }
{ measur(2081) correl(1212) valu(896) }
{ bind(1733) structur(1185) ligand(1036) }
{ network(2748) neural(1063) input(814) }
{ take(945) account(800) differ(722) }
{ studi(2440) review(1878) systemat(933) }
{ motion(1329) object(1292) video(1091) }
{ assess(1506) score(1403) qualiti(1306) }
{ treatment(1704) effect(941) patient(846) }
{ framework(1458) process(801) describ(734) }
{ concept(1167) ontolog(924) domain(897) }
{ clinic(1479) use(1117) guidelin(835) }
{ algorithm(1844) comput(1787) effici(935) }
{ extract(1171) text(1153) clinic(932) }
{ method(1557) propos(1049) approach(1037) }
{ data(1714) softwar(1251) tool(1186) }
{ care(1570) inform(1187) nurs(1089) }
{ general(901) number(790) one(736) }
{ method(984) reconstruct(947) comput(926) }
{ search(2224) databas(1162) retriev(909) }
{ featur(1941) imag(1645) propos(1176) }
{ case(1353) use(1143) diagnosi(1136) }
{ risk(3053) factor(974) diseas(938) }
{ research(1085) discuss(1038) issu(1018) }
{ system(1050) medic(1026) inform(1018) }
{ model(2341) predict(2261) use(1141) }
{ visual(1396) interact(850) tool(830) }
{ compound(1573) activ(1297) structur(1058) }
{ studi(1119) effect(1106) posit(819) }
{ spatial(1525) area(1432) region(1030) }
{ record(1888) medic(1808) patient(1693) }
{ health(3367) inform(1360) care(1135) }
{ model(3480) simul(1196) paramet(876) }
{ ehr(2073) health(1662) electron(1139) }
{ state(1844) use(1261) util(961) }
{ research(1218) medic(880) student(794) }
{ patient(2837) hospit(1953) medic(668) }
{ model(2656) set(1616) predict(1553) }
{ data(2317) use(1299) case(1017) }
{ age(1611) year(1155) adult(843) }
{ medic(1828) order(1363) alert(1069) }
{ signal(2180) analysi(812) frequenc(800) }
{ cost(1906) reduc(1198) effect(832) }
{ group(2977) signific(1463) compar(1072) }
{ sampl(1606) size(1419) use(1276) }
{ intervent(3218) particip(2042) group(1664) }
{ activ(1138) subject(705) human(624) }
{ patient(1821) servic(1111) care(1106) }
{ use(2086) technolog(871) perceiv(783) }
{ analysi(2126) use(1163) compon(1037) }
{ health(1844) social(1437) communiti(874) }
{ high(1669) rate(1365) level(1280) }
{ use(976) code(926) identifi(902) }
{ use(1733) differ(960) four(931) }
{ drug(1928) target(777) effect(648) }
{ implement(1333) system(1263) develop(1122) }
{ estim(2440) model(1874) function(577) }
{ decis(3086) make(1611) patient(1517) }
{ process(1125) use(805) approach(778) }
{ method(1969) cluster(1462) data(1082) }
{ detect(2391) sensit(1101) algorithm(908) }

Resumo

JECTIVES: To obtain a balanced view on the role and place of expert knowledge and learning methods in building Bayesian networks for medical image interpretation.METHODS AND MATERIALS: The interpretation of mammograms was selected as the example medical image interpretation problem. Medical image interpretation has its own common standards and procedures. The impact of these on two complementary methods for Bayesian network construction was explored. Firstly, methods for the discretisation of continuous features were investigated, yielding multinomial distributions that were compared to the original Gaussian probabilistic parameters of the network. Secondly, the structure of a manually constructed Bayesian network was tested by structure learning from image data. The image data used for the research came from screening mammographic examinations of 795 patients, of whom 344 were cancerous.RESULTS: The experimental results show that there is an interesting interplay of machine learning results and background knowledge in medical image interpretation. Networks with discretised data lead to better classification performance (increase in the detected cancers of up to 11.7%), easier interpretation, and a better fit to the data in comparison to the expert-based Bayesian network with Gaussian probabilistic parameters. Gaussian probability distributions are often used in medical image interpretation because of the continuous nature of many of the image features. The structures learnt supported many of the expert-originated relationships but also revealed some novel relationships between the mammographic features. Using discretised features and performing structure learning on the mammographic data has further improved the cancer detection performance of up to 17% compared to the manually constructed Bayesian network model.CONCLUSION: Finding the right balance between expert knowledge and data-derived knowledge, both at the level of network structure and parameters, is key to using Bayesian networks for medical image interpretation. A balanced approach to building Bayesian networks for image interpretation yields more accurate and understandable Bayesian network models.

Resumo Limpo

jectiv obtain balanc view role place expert knowledg learn method build bayesian network medic imag interpretationmethod materi interpret mammogram select exampl medic imag interpret problem medic imag interpret common standard procedur impact two complementari method bayesian network construct explor first method discretis continu featur investig yield multinomi distribut compar origin gaussian probabilist paramet network second structur manual construct bayesian network test structur learn imag data imag data use research came screen mammograph examin patient cancerousresult experiment result show interest interplay machin learn result background knowledg medic imag interpret network discretis data lead better classif perform increas detect cancer easier interpret better fit data comparison expertbas bayesian network gaussian probabilist paramet gaussian probabl distribut often use medic imag interpret continu natur mani imag featur structur learnt support mani expertorigin relationship also reveal novel relationship mammograph featur use discretis featur perform structur learn mammograph data improv cancer detect perform compar manual construct bayesian network modelconclus find right balanc expert knowledg dataderiv knowledg level network structur paramet key use bayesian network medic imag interpret balanc approach build bayesian network imag interpret yield accur understand bayesian network model

Resumos Similares

IEEE Trans Pattern Anal Mach Intell - Are Gibbs-Type Priors the Most Natural Generalization of the Dirichlet Process? ( 0,771809408034747 )
AMIA Annu Symp Proc - Testing the calibration of classification models from first principles. ( 0,754332699728665 )
Med Decis Making - Calibration of complex models through Bayesian evidence synthesis: a demonstration and tutorial. ( 0,74917664069521 )
Neural Comput - A semiparametric Bayesian model for detecting synchrony among multiple neurons. ( 0,744712056317088 )
Lifetime Data Anal - Bayesian nonparametric models for ranked set sampling. ( 0,74382648546658 )
Comput Math Methods Med - Bayesian inference of the Weibull model based on interval-censored survival data. ( 0,742937159972078 )
IEEE Trans Pattern Anal Mach Intell - Modeling Natural Images Using Gated MRFs. ( 0,739443037693338 )
Res Synth Methods - Random-effects meta-analysis of time-to-event data using the expectation-maximisation algorithm and shrinkage estimators. ( 0,73492301536698 )
Comput Math Methods Med - Bayesian hierarchical modeling for categorical longitudinal data from sedation measurements. ( 0,733410270585298 )
Res Synth Methods - A Bayesian nonparametric meta-analysis model. ( 0,728904714851903 )
Spat Spatiotemporal Epidemiol - Goodness-of-fit measures for individual-level models of infectious disease in a Bayesian framework. ( 0,724651659374932 )
Med Decis Making - Bayesian calibration of a natural history model with application to a population model for colorectal cancer. ( 0,721360978791064 )
J. Comput. Biol. - Expectation-maximization algorithm for determining natural selection of Y-linked genes through two-sex branching processes. ( 0,72078597957782 )
J. Comput. Biol. - A spatial haplotype copying model with applications to genotype imputation. ( 0,71943927163176 )
IEEE Trans Image Process - Bayesian robust principal component analysis. ( 0,71884663563538 )
Comput Math Methods Med - An empirical Bayes optimal discovery procedure based on semiparametric hierarchical mixture models. ( 0,716891361327435 )
J. Comput. Biol. - Bayesian blind source separation for data with network structure. ( 0,715247472815124 )
Lifetime Data Anal - Bayesian semiparametric modeling for stochastic precedence, with applications in epidemiology and survival analysis. ( 0,712349331703309 )
Med Decis Making - Assessing uncertainties surrounding combined endpoints for use in economic models. ( 0,71209330537616 )
Comput Math Methods Med - Inference for ecological dynamical systems: a case study of two endemic diseases. ( 0,706748107354902 )
IEEE Trans Image Process - A Bayesian framework for image segmentation with spatially varying mixtures. ( 0,703980716982349 )
IEEE Trans Neural Netw Learn Syst - Variational Bayesian Inference Algorithms for Infinite Relational Model of Network Data. ( 0,697477197893061 )
J. Comput. Biol. - On the inference of dirichlet mixture priors for protein sequence comparison. ( 0,696117498060561 )
IEEE Trans Pattern Anal Mach Intell - Negative Binomial Process Count and Mixture Modeling. ( 0,689505012816015 )
Comput Methods Programs Biomed - Identification of an integrated mathematical model of standard oral glucose tolerance test for characterization of insulin potentiation in health. ( 0,688501349561989 )
J. Comput. Biol. - Characterizing the empirical distribution of prokaryotic genome n-mers in the presence of nullomers. ( 0,688500259332514 )
Res Synth Methods - Bayesian model selection for meta-analysis of diagnostic test accuracy data: Application to Ddimer for deep vein thrombosis. ( 0,687274051799345 )
Spat Spatiotemporal Epidemiol - A Bayesian space-time model for discrete spread processes on a lattice. ( 0,686061426865517 )
Comput. Biol. Med. - Expectation-maximization technique for fibro-glandular discs detection in mammography images. ( 0,685878345246161 )
Lifetime Data Anal - Bayesian local influence for survival models. ( 0,681100415318106 )
IEEE Trans Image Process - Variational Bayesian method for Retinex. ( 0,6758968571473 )
Neural Comput - Learning coefficient of generalization error in Bayesian estimation and vandermonde matrix-type singularity. ( 0,675526902784443 )
IEEE Trans Image Process - Blind image quality assessment: a natural scene statistics approach in the DCT domain. ( 0,673119456063259 )
Med Decis Making - Identifying best-fitting inputs in health-economic model calibration: a Pareto frontier approach. ( 0,670863124963952 )
IEEE Trans Image Process - Generative Bayesian image super resolution with natural image prior. ( 0,669017139039196 )
IEEE Trans Neural Netw Learn Syst - Incorporating Wind Power Forecast Uncertainties Into Stochastic Unit Commitment Using Neural Network-Based Prediction Intervals. ( 0,668749959937437 )
IEEE Trans Image Process - Probabilistic image modeling with an extended chain graph for human activity recognition and image segmentation. ( 0,664664030777598 )
J. Comput. Biol. - Computational methods for a class of network models. ( 0,662264479412034 )
Brief. Bioinformatics - Semiparametric prognosis models in genomic studies. ( 0,659174205867125 )
Res Synth Methods - Critical interpretation of Cochran's Q test depends on power and prior assumptions about heterogeneity. ( 0,657589083598807 )
IEEE J Biomed Health Inform - Sparsity-inspired nonparametric probability characterization for radio propagation in body area networks. ( 0,657571353568579 )
IEEE Trans Neural Netw Learn Syst - Robust Novelty Detection via Worst Case CVaR Minimization. ( 0,657232374054453 )
J Biomed Inform - Link-topic model for biomedical abbreviation disambiguation. ( 0,655244419870289 )
Spat Spatiotemporal Epidemiol - Bayesian hierarchical modeling of the dynamics of spatio-temporal influenza season outbreaks. ( 0,654224165908948 )
IEEE Trans Image Process - Bayesian inference of models and hyperparameters for robust optical-flow estimation. ( 0,649676854419759 )
IEEE Trans Image Process - Computationally tractable stochastic image modeling based on symmetric Markov mesh random fields. ( 0,64814447050216 )
Brief. Bioinformatics - CaliBayes and BASIS: integrated tools for the calibration, simulation and storage of biological simulation models. ( 0,647060765030762 )
Neural Comput - Multiple tests based on a gaussian approximation of the unitary events method with delayed coincidence count. ( 0,646412549146372 )
J Biomed Inform - Learning patient-specific predictive models from clinical data. ( 0,645925339280082 )
IEEE Trans Image Process - Blind separation of time/position varying mixtures. ( 0,64473781118281 )
Med Decis Making - Not simply more of the same: distinguishing between patient heterogeneity and parameter uncertainty. ( 0,644531420206949 )
Med Decis Making - Accounting for methodological, structural, and parameter uncertainty in decision-analytic models: a practical guide. ( 0,642235574223243 )
Brief. Bioinformatics - The dilemma of choosing the ideal permutation strategy while estimating statistical significance of genome-wide enrichment. ( 0,641971776457136 )
Neural Comput - Bayesian sparse partial least squares. ( 0,640790514794132 )
IEEE Trans Neural Netw Learn Syst - Generalized multiple kernel learning with data-dependent priors. ( 0,635769205990333 )
Spat Spatiotemporal Epidemiol - Inference from ecological models: estimating the relative risk of stroke from air pollution exposure using small area data. ( 0,635766444159321 )
Comput Methods Programs Biomed - Multivariate Bayesian modeling of known and unknown causes of events--an application to biosurveillance. ( 0,63419680125501 )
IEEE Trans Pattern Anal Mach Intell - Articulated Human Detection with Flexible Mixtures-of-Parts. ( 0,6337136842723 )
Res Synth Methods - A basic introduction to fixed-effect and random-effects models for meta-analysis. ( 0,633567515762687 )
BMC Med Inform Decis Mak - A simulation model of colorectal cancer surveillance and recurrence. ( 0,6332678046844 )
Comput Methods Programs Biomed - NIMROD: a program for inference via a normal approximation of the posterior in models with random effects based on ordinary differential equations. ( 0,628988415585406 )
Neural Comput - Determination and the no-free-lunch paradox. ( 0,627879865852286 )
Comput Methods Programs Biomed - The exponentiated exponential mixture and non-mixture cure rate model in the presence of covariates. ( 0,627429119592494 )
Neural Comput - Efficient Markov chain Monte Carlo methods for decoding neural spike trains. ( 0,626035626228092 )
Comput Math Methods Med - Applications of Bayesian gene selection and classification with mixtures of generalized singular g-priors. ( 0,625289063588576 )
Neural Comput - Efficient sensory encoding and Bayesian inference with heterogeneous neural populations. ( 0,623969640291389 )
J. Comput. Biol. - NP-MuScL: unsupervised global prediction of interaction networks from multiple data sources. ( 0,623297972617628 )
Spat Spatiotemporal Epidemiol - A multivariate CAR model for mismatched lattices. ( 0,622207409986834 )
Med Biol Eng Comput - A poisson process model for hip fracture risk. ( 0,620707887011588 )
IEEE Trans Pattern Anal Mach Intell - Bayesian Nonparametric Methods for Partially-Observable Reinforcement Learning. ( 0,620164219007019 )
Med Decis Making - A framework for addressing structural uncertainty in decision models. ( 0,619101906669138 )
J. Comput. Biol. - Continuous-time Markov chain-based flux analysis in metabolism. ( 0,617389023152753 )
Artif Intell Med - Impact of precision of Bayesian network parameters on accuracy of medical diagnostic systems. ( 0,615774300479027 )
AMIA Annu Symp Proc - Learning to predict post-hospitalization VTE risk from EHR data. ( 0,615288494913394 )
Spat Spatiotemporal Epidemiol - The detection of spatially localised outbreaks in campylobacteriosis notification data. ( 0,615261528106121 )
IEEE Trans Pattern Anal Mach Intell - Gaussian Process-Mixture Conditional Heteroscedasticity. ( 0,612073026472261 )
IEEE Trans Image Process - Posterior-mean super-resolution with a causal Gaussian Markov random field prior. ( 0,611774482973929 )
Neural Comput - The neural representation of time: an information-theoretic perspective. ( 0,610217783395624 )
J. Comput. Biol. - A Bayesian sampler for optimization of protein domain hierarchies. ( 0,609058784318225 )
Neural Comput - Attention as reward-driven optimization of sensory processing. ( 0,608581093717939 )
IEEE Trans Image Process - A study of multiplicative watermark detection in the contourlet domain using alpha-stable distributions. ( 0,608004092202914 )
Artif Intell Med - Improving Bayesian credibility intervals for classifier error rates using maximum entropy empirical priors. ( 0,607992676574455 )
Med Decis Making - Comparing Bayesian and frequentist approaches for multiple outcome mixed treatment comparisons. ( 0,607863582197788 )
Comput Methods Programs Biomed - A Bayesian multilevel model for fMRI data analysis. ( 0,606496311394632 )
J. Comput. Biol. - Exploiting genome structure in association analysis. ( 0,606084389989766 )
J. Comput. Biol. - Maximum parsimony, substitution model, and probability phylogenetic trees. ( 0,604977169261852 )
Comput. Biol. Med. - A pattern-oriented specification of gene network inference processes. ( 0,604627356988756 )
IEEE Trans Image Process - Bayesian nonparametric dictionary learning for compressed sensing MRI. ( 0,602314810123059 )
Med Decis Making - Linear regression metamodeling as a tool to summarize and present simulation model results. ( 0,601636050235488 )
IEEE Trans Pattern Anal Mach Intell - Background Subtraction with Dirichlet Process Mixture Models. ( 0,599123412692547 )
Med Decis Making - The combined analysis of uncertainty and patient heterogeneity in medical decision models. ( 0,597422927231007 )
Comput Methods Programs Biomed - Anticipating the next meal using meal behavioral profiles: a hybrid model-based stochastic predictive control algorithm for T1DM. ( 0,597372919911948 )
Spat Spatiotemporal Epidemiol - A space-time point process model for analyzing and predicting case patterns of diarrheal disease in northwestern Ecuador. ( 0,596016233524222 )
Neural Comput - Universal approximation depth and errors of narrow belief networks with discrete units. ( 0,59547118947939 )
IEEE Trans Image Process - Shape-based normalized cuts using spectral relaxation for biomedical segmentation. ( 0,594754455249742 )
IEEE Trans Image Process - Counting people with low-level features and Bayesian regression. ( 0,592195052563939 )
IEEE Trans Pattern Anal Mach Intell - Temporal Analysis of Motif Mixtures using Dirichlet Processes. ( 0,590794771082532 )
Comput Math Methods Med - Immune response to a variable pathogen: a stochastic model with two interlocked Darwinian entities. ( 0,586708046434557 )
Comput Methods Programs Biomed - Life prediction of different commercial dental implants as influence by uncertainties in their fatigue material properties and loading conditions. ( 0,586352729855949 )
BMC Med Inform Decis Mak - A novel differential diagnostic model based on multiple biological parameters for immunoglobulin A nephropathy. ( 0,586313639304403 )