Spatial-temporal regulation method for robust model estimation
View Patent ↗A computerized spatial-temporal regulation method for accurate spatial-temporal model estimation receives a spatial temporal sequence containing object confidence mask. A spatial-temporal weight regulation is performed to generate weight sequence output. A weighted model estimation is performed using the spatial temporal sequence and the weight sequence to generate at least one model parameter output. An iterative weight update is performed to generate weight sequence output. A weighted model estimation is performed to generate estimation result output. A stopping criteria is checked and the next iteration iterative weight update and weighted model estimation is performed until the stopping criteria is met. A model estimation is performed to generate model parameter output. An outlier data identification is performed to generate outlier data output. A spatial-temporal data integrity check is performed and the outlier data is disqualified.
1. A computerized spatial-temporal regulation method for more accurate spatial-temporal model estimation in quantitative microscopy data analysis for detecting puncta related phenotypes comprising the steps of:
a) Inputting a spatial temporal sequence containing an object confidence mask from microscopy time-lapse studies of subcellular components;
b) Using the spatial temporal sequence to generate a temporal confidence sequence from variation of adjacent frame temporal intensity and integrating with the object confidence mask to obtain a weight sequence output;
c) Performing weighted fitting using the spatial temporal sequence and the weight sequence to obtain at least one model parameter output.
2. The spatial-temporal regulation method of claim 1 wherein the temporal confidence sequence T i (x,y) is generated for image pixel of 8 bit deep is defined by
T i ( x,y )=255−(255×Δ F i ( x,y ))
where ΔF i (x,y) is variation of adjacent frame temporal intensity.
3. The spatial-temporal regulation method of claim 1 wherein the weighted model estimation uses a modified nonlinear regression method.
4. The spatial-temporal regulation method of claim 1 wherein the at least one model parameter is a destaining constant τ wherein τ is 1/α of destaining model I(t i )=Ae αt i +B.
5. A computerized iterative improvement method for more accurate spatial-temporal model estimation in quantitative microscopy data analysis for detecting puncta related phenotypes comprising the steps of:
a) Inputting a spatial temporal sequence containing an object confidence mask and an initial weight sequence from microscopy time-lapse studies of subcellular components;
b) Performing an iterative weight update to generate a weight sequence output;
c) Performing a weighted fitting using the spatial temporal sequence and the weight sequence to obtain an estimation result output;
Performing a stopping criteria check and continuing the next iteration of iterative weight update and weighted model estimation until the stopping criteria is met;
Outputting the estimation result as the model parameter output when the stopping criteria is met.
6. The iterative improvement method of claim 5 wherein the iterative weight update method uses an exponential function to discriminate higher confidence from lower one.
7. The iterative improvement method of claim 5 wherein the iterative weight update method uses errors between the input spatial temporal sequence and expected sequence based on the estimation result to generate weight sequence.
8. The iterative improvement method of claim 5 uses the spatial temporal sequence to generate a temporal confidence sequence from variation of adjacent frame temporal intensity and integrating with the object confidence mask to obtain a weight sequence output.
9. The iterative improvement method of claim 8 wherein the temporal confidence sequence T i (x,y) is generated for image pixel of 8 bit deep is defined by
T i ( x,y )=255−(255 ×ΔF i ( x,y ))
where ΔF i (x,y) is variation of adjacent frame temporal intensity.
10. The iterative improvement method of claim 5 wherein the model parameter is a destaining constant τ wherein τ is 1/α of destaining model I(t i )=Ae −αt i +B.