ResearchGate Q&A lets scientists and researchers exchange questions and answers relating to their research expertise, including areas such as techniques and methodologies.
Browse by research topic to find out what others in your field are discussing.
Objective evaluation of segmented Image?
I need to evaluate segmented image objectively. Can you help me find MATLAB Code for any Objective evaluation of segmentation algorithm?
Assuming that you have a gold standard for the segmentation (manual segmentation or reference algorithm), there are different measures of similarity that could be used to measure the effectiveness of your segmentation algorithm. None of them is superior to the others, it strongly depends on your goal and on the application.
This nice paper cover some of the measures of similarity that can be used and easily implemented in Matlab http://www.ri.cmu.edu/pub_files/pub4/unnikrishnan_ranjith_2007_1/unnikrishnan_ranjith_2007_1.pdfFollowing
Determining linearity between the dependent and independent variable?
I have a dependent variable that is continuous. Another independent variable that is categorical (presence or absence). Thus based on the presence represented as1 and absence as 0, the dependent variable has certain value.
Before one builds a linear regression model, if we wish to find out one of the most imp. assumptions that whether the two variables are linear....it is advised to generate a scatter plot.
I assume that if both the variables (dependent & independent) are continuous, then its fine.
But if the independent has only 2 values (0 & 1) as in my case, the scatter plot would be just 2 straight lines, one for each value based on the values for the dependent variables.
How does one check for linearity between such cases?
Linearity can only be tested if we have at least three points. I assume you are therefore trying to test whether the assumption of linearity of the association between the continuous variable and your independent variable is met.
Why favour entering a dependent variable as a continuous variable rather than entering a dichotomised variable?
If an association is dose dependant, conserving information on the value of a measured factor will improve our ability to model it's association to the independent variable. In other words, the analysis will also test whether the association is "dose-dependent" and thereby provide support towards a causal relationship.
Why not just create dummy variables for different ranges of values instead?
This approach indeed has the advantage of not having to assume that the magnitude of the association is always proportional to the value of the dependent variable. It however has a major disadvantage! When creating ranges of values, we generate what we call dummy variables. Each new variable then corresponds to one of the ranges (n= number of ranges). Therefore, we are using n-1 degree of freedom in the model. This considerably reduces the power of the overall analysis especially if some ranges only contain very few participants. For this reason, we favour testing the assumption of linearity.
What are we testing?
When entering a continuous variable in logistic regression, the reported indice from the logistic regression corresponds to the average logged odd ratio for each unit of the independent variable. For this to make sense, it is necessary to verify that it remains more or less constant over the entire scale.
Generating dummy variables
This can be done by first transforming the continuous variable into an ordinal one. The number of ranges should depend of the sample size. The more the better, but each range should contain enough participants to have about 10 participants with the outcome of interest. First construct the ranges with similar paces between each range.
Chose the reference value and code it dummy_0. Code the other ranges by units from then (negatively bellow and positively above). Compute the logistic regression model entering each range as a separate variable (keep out the reference variable). Visually analyse the magnitude of association of each variable and check if it seems linear. If this is not the case, consider transformation and start the operation again.
Testing linearity – likelihood ratio test
The idea is to see whether we have lost information when considering the association to be linear. Therefore, what we do is to compare the model with dummy variables with the model with the same measure entered as a continuous variable. If the models significantly differ, this means we cannot assume the association to be linear. Therefore, we perform a likelihood ratio test between models and if p>0.05 then we can state that absence of linearity was not observed.
Stats automatically generates dummy variables using the xi: command.
gen dependant2=dependant // generates a new variable
recode dependent2 0/0.999=0 1/1.9999=1 2/2.9999=2 3/3.9999=3 4/4.9999=4
xi: logit independant i.dependant2, or
xi: logit independant dependant, or
lrtest A BFollowing
What are differences between conservation and preservation of natural resources?
The goals of these two terminologies differ not only with respect to their explicit
and implicit long-term objectives, but also with respect to their justifications,
their immediate targets and obstacles, and the strategies that are likely to achieve these targets
Conservation means to reduce the usage of natural resources, to use natural resource more efficiently; in particular, non-renewable resources.
Preservation means to protect or save natural resources in the present for the purpose of using them in the future. If exploitation occurs at a future date, the basis of protecting the resources today is so that future generations are able to benefit from them under better conditions.Following
Can anyone give references to Talmy’s works on Figure/Ground?
Cognitive Linguistics embarked on a radical course, one that placed conceptual analysis and cognitive principles squarely at the forefront of the study of mind and language.
For a more systemic and thorough explanation, see Talmy's two volume Toward a Cognitive Semantics, especially Chapter 5 of the first volume.Following
Is it possible to have light without electrical power?
In general, light is produced for household purpose with bulb, CFL, tube, LED etc from electrical power. There is a big portion of world that lives in dark after sun sets. These people do not want electrical power, they want light first.
Is it possible to create a pollution free light without electric power and without burning something. Light that is sufficient to illuminate a 12 x 12 feet hut without an electricity connection. Light that is very very cheap (not more than $1 per week and 8 hours per day)….Don’t tell solar or known things….suggest something wild…some new idea….innovative…no power only light….Just for the sake of discussion….
Naveen, one thing can be fake for one person and fully functional for another. One paper can be brilliant for one researcher and may worth nothing for the other one. Please search deeply on the subject, download all patents, papers and videos. At least, this will make you a good idea about the subject and will test your ability for doing a complete documentation. Try to understand which is the original idea and which are fake or useless copies. For example an useless one: http://jnaudin.free.fr/kapagen/k32.gif
The idea of extracting energy during a resonant process (without killing the resonance) is not new. The paper you have uploaded is interesting anyway, I've read it long time ago.Following
Can anyone knows why energy,entropy,skewness,difference mean and Kurtosis are chosen as a features for classification?
I am vigneshwaran PG student doing a research in medical thermography.Can anyone knows why energy,entropy,skewness,difference mean and Kurtosis are chosen as a features in classification of normal and abnormal patterns in breast thermograms?If any one knows the answer to my question please answer
thank you sir for your answersFollowing
Can anyone advise me on solving the cloning problem?
I inserted a 3131bp fragment (PCR product) to a vector backbone with 3600bp length by two enzymes ECORI and BSP. After plasmid extraction and linearization, the vector run on the Agarose gel. I must be observed a 6731bp band on gel for linearized plasmid, but there is a 4500 bp band.
Foe sequencing of 3131bp fragment (PCR product) in plasmid, I designed 5 primers.
Sequencing for first primer is positive and for others is failed.
This data and checking of plasmid with restriction enzymes showed that only 900 bp of 3131bp fragment (PCR product) is cloned to vector.
Is it possible occurrence of such event?Following
Is that any effect of tool-chip contact length on crater wear during turning?
Also, what is the effect of speed, feed on tool-chip contact length?
Does anybody know something about opportunistic behaviour and student performance in online or non presential activities?
I have seen this type of opportunistic behavior among students in several online activities and I would like to know more about it and its effect on the final grade. I would like to know if there is seminal literature on that topic focus on Higher Education, especially in economics, business and financial accounting. Thanks!Following
How do I get error bars in a correlation graph in GraphPad Prism 6?
Comparing XY data, for each data (X and Y) I have the mean value, SEM and N. I want to show the correlation between the groups. How can I illustrate error bars in a correlation graph in Prism 6. I only get the correlation graph and can´t activate the "show error bars" buttom in "format graph". Thanks!Following
How can I prepare Nano cellulose in a dry form?
I want to do physical and chemical characterization for my prepared nanocellulose sample such as IR analysis, X-ray diffraction,degree of oxidation ... etc.So, do air or oven dry effect the sample properties. and if so what is the alternate way to dry it without loosing its properties.
Dear Dr. Abdellatif Barakat
Thank you so much for your valuable add. but I still have a problem in drying as Freeze drying is not available at my lab unfortunately.Following
What are the necessary procedures to check the outputs of education, learning accounting?
It is known that there are many of the expected outputs of accounting education program, including for example:
Knowledge, practical skills, social skills, and other
This requires the existence of a mechanism which is to make sure these requirements are met.
Thank you Armando Fernández GuillermetFollowing
Anyone familiar with diazonium coupling of silk?
I have done diaotization of light chain silk for modification of tyrosine present in it. I tried H1-NMR and ATR-FTIR on these modified silk. But no peak was observed, from dizonium salts in both of these characterization, Is there any technique to determine that silk material is diotized and %diaotization done after completion of experiment?
Thank you all for your kind suggestionFollowing
How can I choose a training set from database ?
in Face recognition using PCA ,
how we can choose the training sets and testing sets from given database ??
and what is the minimum and maximum number of these sets if we are working on At&T database ??
In case of At&T database use "leave-one-out cross validation" to separate training and test set.Following
How far can one get with analogies in theoretical physics?
Many a great theoretical physicist have gone on to explain their work qualitatively to the public, often times by virtue of analogies.
Now, I know analogies and qualitative explanations can be misleading if not presented correctly. Having a little background in mathematics (basic ideas in tensor descriptions of space-time etc.) I realize some of these analogies presented to me in the past are very misleading, while others are almost as good as a conceptual "one-to-one mapping" (so that understanding of the analogy leads to an almost perfect understanding of the concept)... almost as if the analogy belongs to the same abstract category as the concept to be explained.
I would just like to know (though I guess there is no exact answer): how far can one get to a full understanding of theoretical topics with qualitative explanations. Or, on a more pessimistic note: is it worth explaining things like general relativity, quantum mechanics, and string theory qualitatively and by analogy for any other purposes than "sensational" science?
Petrus> ... bit skeptical of your approach
What did you interprete to be my approach?Following
Lognormal Size Distribution and Multimodal Size Distribution by Dynamic Light Scattering (DLS) measurement ?
The results of DLS measurment always give us Lognormal Size Distribution and Multimodal Size Distribution. I do not know exactly what is the different between them? and Which one is better?. For my experiment, I want to survey the change of diameter of nanoparticle after adsorption an extra molecular.
Could everybody give me some advises?
Many thankful for !Following
What is a suitable way to calculate the reference range of a protein in plasma?
I would like to estimate the reference range of a protein in plasma, the mean and deviation standard were: 152 and 58 µg/ml, respectively. The common formula to calculate the reference range is: mean±1.96x SD. The problem is that the SD in my control is to high and the estimated reference range is too large, can I use the confidence interval of the mean as a reference range instead of the common formula?
Ramnath is on the right way:
The calculation as Mean±1.96xSD should give the range of values covering about 95% of the values that one may expect. This means: the reference range is calculated so that from really many many measured values about 95% of them will fall within this range, 2.5% will be lower than the lower limit, and 2.5% will be larger than the upper limit.
The calculation is based on a model about the distribution of such values, and this model is known as the "normal distribution" or Gaussian distribution. This model may or may not be sensible for a particular case. One needs to check this. If you have a lot of data (some 100s or better 1000s of values) then you can for instance plot a histrogram and compare the shape to the shape of a normal distribution (with mean and variance calculated from the data). If you have less data you can use a normal-quantile-quantile plot to check if the Gaussian distribution model is adequate (see http://en.wikipedia.org/wiki/Q%E2%80%93Q_plot).
In your case the Gaussian model is likely inadequate, because concentrations of biological molecules usually do not follow a Gaussian distribution. Their distribution is often right-skewed. Examples:
Sometimes the logarithms of such right.skewed values have a distribution that is approximately Gaussian. So you can calculate the reference range for the logarithms. The (lower and upper) limits of this range can be simply antiloged again to get the range on the original scale. This is what Ramnath proposed.
Say your values are called "y" (y1, y2, y3, ...)
Then first calculate the logs: y' = log(y)
Now check again if the distribution of the y' values is approximately gaussian (histogram or QQ-plot).
If this is the case, you calculate the mean and the SD of these y' values.
The reference range for y' is then mean(y') ±1.96xSD(y'):
the lower limit is L' = mean(y') -1.96xSD(y'):
the upper limit is U' = mean(y') +1.96xSD(y'):
The range for y is obtained by antiloging these limits:
L = exp(L')
U = exp(U')
This is it.
If the distribution of y' is not approximately gaussian, then you need to claculate the appropriate quantiles directly from the data you have. This will give you possibly quite bad estimates for the reference range when you have only few values. If you have some 100s or 1000s of values, this method works good and always - independent of the actual distribution of the values.
The lower limit of the range is simply the 0.025-quantile of the values,
the upper limit of the range is simply the 0.975-quantile of the values.
You can use a statistics software (even Excel will do!) to get these quantiles. You can alternatively sort all the values and use the 2.5%-smallest and 97.5%-smallest value as the limits (these are the searched quantiles). Example: if you have 1000 values in a sorted list (from small to large), the 0.025*1000 = 25th value will be the value of the lower limit and the (1000-25)=975th value in this list will be the upper limit.Following
Can anyone give me the correct time for measuring of absorbance after adding buffers (pH1 & pH4.5) in determination of total Anthocyanin?
I am conducting an experiment to quantification of total anthocyanin in Tea (Camellia sinensis) using spectrophotometric method. I took absorbance at 520nm and 700nm with two buffers (pH=1 and pH=4.5) as formula mentioned in Journal of AOAC International vol. 88:5 1269-1278 (Lee et al, 2005) in determination of total Anthocyanin.
According to formula Absorbance (A)= (A520nm – A700nm)at pH 1.0 - (A520nm – A700nm)at pH 4.5
Literature suggested to take absorbance within 20-50 min of preparation. But absorbance is increasing continuously with the time (even less than 20min) at both 520 nm & 700nm only in pH=4.5 buffer. As a result of higher absorbance values, final value become negative (difference between Abs at pH=1 and pH=4.5). Please anybody explain this incident or suggest any other spectrophotometric method for quantification of Anthocyanin in Tea.
How to create a composition of different wavelength (channel) images in Matlab?
I would like to create a color image from a set of intensity images (grayscale) each one corresponding to different wavelengths (colors). In order to do that, for each image:
- I set the wavelength to the hue (H) value (from red=0 to blue=2/3) thanks to a simple linear transformation
- I set saturation (S) to 1 (fully saturated colors)
- I normalize the intensity and set it to the value (V)
Although this part seems to work properly, when I try to combine the corresponding set of HSV images, I am getting strangely colored results.
So far, the most coherent results have been achieved for imfuse with the alpha blending option after transforming the HSV images to RGB but I would expect different outcome. (see http://stackoverflow.com/questions/25587530/fusing-more-than-2-images-in-matlab). I have also tried changing the weight factor to 1/N where N is the number of images.
Can you suggest some other way of forming a single colored image from the different channels (>3)?
it looks like you are dealing with two degree of freedoms: wavelength and intensity. Therefore, setting Saturation of 1 is correct.
Now, for each pixel value, I would sum up the 2D vector (HV) describing the Hue and V for that particular pixel/channel combination (in the conical HSV space, being on the border border of the cone as S = 1). You can then convert the resulting HSV image to the RGB space to visualise it.
Depending on your application and desired results, you can eventually visualise the norm of the resulting vector, or only the resulting direction (H) or any other combination.
I hope this will be helpful for you.Following
I used Kruskal-Wallis and thinking of using Dunn's multiple comparison test as post hoc. Is this correct?
I have a set of data with 3 variables. Each variable has either 4 or 6 measurements - fluorescence intensity. The variables are independent. I checked if the data followed a normal distribution and it doesn't. The data is all numerical. Based on the facts that I have nonparametric, numerical data with more than 2 variables I decided to use the Kruskal Wallis to determine if the differences observed in fluorescence measurements are different for the 3 variables. I do obtain a list with p-values smaller than 0.05 therefore there are statistically different fluorescence values. Now I would like to know which one of the 3 variables is the one that is different. For that I believe I will have to perform a post hoc test and online I found a website (http://media3.bmth.ac.uk/spss/focus_pages/focus_13a.htm) saying that the Dunn's multiple comparison test is the one to use when sample sizes are not equal (I have either 4 or 6 sample size for a variable). Is this the correct way to proceed?
Meaning did I chose the right statistical test (Kruskal Wallis) and if so is Dunn's multiple comparison test appropriate as a post hoc?
In terms of " I regularly transform the raw data into normal quantiles first. This is referred to as Normal Scores Test. Van der Waerden suggested such an approach many years ago, and it is available in SAS under PROC RANK. I prefer the option BLOM. "
Is it possible for you to send me one of your applications?
I would be delighted.
A protocol for catalase and superoxide dismutase assay?
I am planning to assay catalase and superoxide dismutase in tissues. Currently I was using the Cayman's kits; however, I have now a lot of samples to assay SOD and CAT and would rather prepare the reagents on my own. Does someone have a protocol for that and could share it with me? I can use either a microplate reader or a spectrophotometer for 1 ml cuvette.
for SOD follow the method of
Beauchamo, C.O. and Fridovich, I. (1971). Superoxide Dismutase: Improved assay and an assay applicable to acrylamide gels. Anal. Biochem. 44: 276-287.Following
How can I resolve the coherent DOA estimation problem?
How to resolve the coherent DOA estimation problem? i need brief answer asap.
I agree with ShahriarFollowing
How do I select the best remote sensing system for monitoring wetland vegetation over a time series?
I have an interest in remote sensing of vegetation health; particularly aquatic vegetation. I would like to evaluate the effects of wetland encroachment on surface water and energy fluxes, to derive an indicator for wetland vegetation health. I want to use data over a time series of about 30 years. I am trying to determine which sensor system is most suitable for my work (Landsat, MODIS or ASTER) as I foresee where data quality might be a limitation given the climatic conditions of the study area (misty for most times of the year). Can anyone with experience in wetland vegetation monitoring give some advice?
Would microwave be the only remote sensing technique I have as an option? Also, is it possible to run imagery from a combination of sensors without risking the credibility of the model? I also want to select a most appropriate ET model, which comes with free and open access.
Thanks in advance for the guidance.
I suggest you that Landsa 5- TM, Landsa7-ETM+ and Landsa 8- OLI data is the bast for vegetation parameter analysis. available from 1987-2014..........DataFollowing
How can I enhance the efficiency of transformation of Ler ecotype with Agrobacterium tumerfaciens?
Recently, I used Agrobacterium tumerfaciens EHA105 strain transformed Ler with the classical flower dipping method. Screened the T1 seeds with antibiotics, only few positive seedlings was harved. Is there any strain or tips can enhance the transformation efficiency?
Rather than flower dipping, one more method is highly suitable .you can try with google search. Positively transformed means not 100% transformation. first you standardize the times for incubation with increase or decrease in time sequence as you have relatively result in positive seedlings some what doughtfull dear. read paper of SL sandaFollowing
English language is the world's Lingua Franca?
Nowadays, it is obvious that English language is a bridge language that is used for almost everything. The question is: Do you think that this is temporary? If yes, which language or languages would be rivaling English in the future, and why?
English rose as a global lingua franca because of two countries, first GB and then US. The rise of the English language was accompanied by the economic, political, and military expansion of these two major host countries, this trend has been continuing for nearly 200 years (I am not sure about the exact duration). Colonialism is another cause for the wide spread of the language, especially in the 19th century. So all in all we have two reasons for the worldwide use of the language: the dominance of the countries using the language (nowadays chiefly US) in the world; the other one is the sheer number of countries using English as their official (national) language, hence huge number of native/L1 speakers. This trend would continue for another long period of time, however, when new world power arises, e.g. China, when the country's economic, political, and military influence expands, the Chinese language would also become a powerful (widely used) language. In view of the global situation, I would assume Chinese would most probably become the next lingua franca (but it would take decades, even centuries to accomplish this). So at least for the time being, learn English (if one is not a native speaker of the language) and try to be a global citizen.Following
How do I identify ions in ESA?
Any idea on how do I identify the passing ions through electrostatic analyser. Signal obtained on a scope out of electron multiplier tube. Using TOF is useful, however difficult to separate ions that have both close delta T and mass to charge ratio.
What kind of ion would you like to detect?Following
How can I optimize pGL3 vector ligation?
I'm having a hard time cloning 2 fragments (2,2kb and 2,3kb) into pGL3-Promoter Vector (5lb), for some time now.
Firstly, in order to linearize my modified pGL3 with a linker containing NdeI and SacII sites and also to get my fragments (which also contain NdeI and SacII sites) out of TOPO vector, I performed a overnight double digestion at 37C with NdeI and SacII (separated tubes, of course), using cutsmart buffer. Next day, the enzymes were inactivated at 65C for 20 min and then dephosphorylation was done by adding to the tubes 1uL of CIP, kept at 37C for 3h.
Both linear dephosphorylated pGL3 and fragments were separated by 0.8% agarose gel, and only the desired bands were taken out from the gel and purified (using Zymo Research DNA recovery kit, using water instead of elution buffer at the final step).
The ligation was performed overnight at 16C using T4 ligase, starting with 100ng of vector and different amounts of insert, such as 1:1, 1:3, 1:5 and 1:10 (vector:insert).
Finally, 50 uL of 5-alpha competent cells were chemically transformed using 5-7uL of the ligation reaction by 30min pre incubation on ice, 42s thermal shock, 250mL of SOC, 1h incubation at 37C shaker and then plated on LB-agar containing ampicillin.
The next day, most plates contained very few colonies, which after verification, none of them had the insert.
Any suggestions on how I can make this work?
since you have isolated the fragments and ligated to the same plasmid my immediate thoughts are that the 2.2kb fragment is the problem.
1. the fragment could be deleterious to the cells. Do the cultures grow slower than expected, are the colonies smaller or take longer to grow than expected, are there tandem repeats in the sequence. There maybe a cryptic promoter in your sequence.
2. I would sequence the clones using plasmid specific primers so that you can see what insert is present and match that to the known sequence.
3. Try transforming into other strains of bacteria. i.e. SURE cells can cope with tandem repeats.
I had a cryptic promoter present in one of my cloning projects and colonies grew very slowly >24hr to get colonies. Perhaps leave plates longer than normal to develop colonies
Should clinical psychologists working in hospitals wear uniforms?
Most clinical psychologists working in hospitals wear their own clothes. Confronted with risk of infection they wear protective clothing.
An increasing number of psychologists are forced to wear staff uniforms.
Should clinical psychologists working in hospitals wear uniforms? Does it affect the therapeutic relation? Looking for literature on this topic.
I would think your personality if you looks neat not need but if not you would be better wear uniform.
Is there anyone who used fructose diet (Altromin C1068) for inducing insuline resistance?
Is there anyone used fructose diet (Altromin C1068) for inducing insuline resistance?
Our group has been using a fructose solution to induce metabolic syndrome-like effects is Sprague-Dawley rats for several years with good results. Food is normal pellet diet. These animals develop mild hypertension in approximately forur weeks or less.Following