Sweep: Turn app into a cli tool about calcium-analysis HOT 8 CLOSED

retroam commented on August 11, 2024 2

Sweep: Turn app into a cli tool

from calcium-analysis.

Comments (8)

sweep-ai commented on August 11, 2024

I'm sorry, but it looks like an error has occured. Try removing and re-adding the sweep label. I'll try again in a minute. If this error persists contact [email protected].

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

I'm sorry, but it looks like an error has occured. Try removing and re-adding the sweep label. I'll try again in a minute. If this error persists contact [email protected].

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

Hey @retroam,

I've started working on turning the app into a CLI tool. The plan is to convert the MATLAB scripts to Python for consistency and ease of packaging. I'll also modify the existing Python scripts to accept command line arguments for input and output directories. Lastly, I'll create a main script that ties all the functions together and serves as the entry point of the CLI tool.

Give me a minute!

Some code snippets I looked at (click to expand). If some file is missing from here, you can mention the path in the ticket description.

Calcium-analysis/README.md

Lines 1 to 2 in c2679e1

 # Calcium-analysis 

 Python implementation of calcium imaging analysis pipeline for cardiac myocytes for [Automated image analysis of cardiac myocyte Ca2+ dynamics](https://pubmed.ncbi.nlm.nih.gov/22255377/)

Calcium-analysis/calcium_analysis/data_extraction.py

Lines 1 to 39 in c2679e1

 import numpy as np 

 from skimage import measure 

 import os 

 def roi_to_data(roi_file, image_file): 

 """ 

  Extract data from regions of interest in an image stack. 

  This function loads a labeled image (from a .npy file), computes 

  the mean intensity of each labeled region in each image in a stack, 

  and saves the results. 

  """ 

 # Load the labeled image and the image stack 

 labels = np.load(roi_file) 

 I_t = np.load(image_file) 

 # Extract data from each region in each image 

 data = [] 

 bkg = [] 

 for I in np.rollaxis(I_t, axis=2): 

 # Get properties of labeled regions 

 props = measure.regionprops(labels, intensity_image=I) 

 # Compute the mean intensity of each region and the background 

 avg = [prop.mean_intensity for prop in props] 

 I_bkg = I.copy() 

 for prop in props: 

 I_bkg[prop.coords[:, 0], prop.coords[:, 1]] = np.nan 

 bkg_intensity = np.nanmean(I_bkg) 

 data.append(avg) 

 bkg.append(bkg_intensity) 

 # Save the results 

 np.save('data.npy', data) 

 np.save('bkg.npy', bkg) 

 # Return the results

Calcium-analysis/tests/test_image_processing.py

Lines 1 to 19 in c2679e1

 import unittest 

 import numpy as np 

 import os 

 from PIL import Image 

 import image_processing # assuming the module is named image_processing.py 

 class TestImageProcessing(unittest.TestCase): 

 def test_load_image(self): 

 # Test loading a single image 

 image_path = 'path_to_a_test_image.tif' 

 I, info = image_processing.load_image(image_path) 

 self.assertIsInstance(I, np.ndarray) 

 self.assertIsInstance(info, dict) 

 self.assertEqual(I.shape[0], info['ImageLength']) 

 self.assertEqual(I.shape[1], info['ImageWidth']) 

 if __name__ == '__main__':

Calcium-analysis/matlab/imagestostacks.m

Lines 1 to 31 in c2679e1

 function[I_t] = imagestostacks(image_fld,save_fld) 

 % IMAGESTOSTACKS(image_fld,save_fld) assembles tif images in image_fld as 

 % a single matrix which is then saved in save_fld as a .mat file 

 % NOTE: tif files in image_fld have to be labelled as image01...09,10 etc. 

 %Example: 

 % I = imagestostacks(image_fld,save_fld); 

 cd(image_fld); 

 i_bloc = dir('*.tif*'); 

 I_t = []; 

 for j = 1:length(i_bloc) 

 info = imfinfo(i_bloc(j).name); 

 fileName{j} = i_bloc(j).name; 

 num_images = numel(info); 

 I_cum = zeros(info(1,1).Height,info(1,1).Width,num_images-1);%pre-alloc 

 disp(['Loading tif file ' num2str(j) ' of ' num2str(length(i_bloc))]); 

 for k = 1:num_images 

 I = imread(i_bloc(j).name, k, 'Info', info); 

 I_cum(:,:,k) = I; 

 end 

 close all force; 

 I_t = cat(3,I_cum,I_t); 

 clear I_cum; 

 end 

 save([save_fld '\I.mat'],'I_t');

Calcium-analysis/matlab/stackstoroi.m

Lines 1 to 41 in c2679e1

 
 function[CC,I_mean,I_bw2,I_overlay] = stackstoroi(save_fld,sd) 

 % STACKSTOROI(image_fld,sd) loads *I*.mat from save_fld for image 

 % segmentation. sd is how many standard deviations above the mean the  

 % threshold is set for segmentation. Outputs: 

 % CC: connected components found in I_bw2 

 % I_mean: averaged image of *I*.mat 

 % I_bw2: binary image containing image segementation of *I*.mat 

 % I_overlay: overlay of identified ROI over I_mean 

 %Example: 

 % [CC,I_mean,I_bw2,I_overlay] = stackstoroi(save_fld,sd); 

 cd(save_fld); 

 m_bloc = dir('*I*.mat*'); 

 I_cum = []; 

 img_content = []; 

 if length(m_bloc)== 11 

 for i = 1:11 

 I_cum = cat(3,I_cum,importdata(m_bloc(i).name)); 

 end 

 else 

 I_cum = importdata(m_bloc(1).name); 

 end 

 for j = 1:length(I_cum) 

 I = I_cum(:,:,j); 

 img_content(j) = mean(I(:)); 

 end 

 [marks,locs] = findpeaks(img_content,'minpeakdistance',100); 

 I_mean = mean(I_cum(:,:,locs),3); 

 %I_bw1 = bwspecial(otsu(I_mean,3)); 

 I_bw1 = threshold(I_mean,mode(I_mean(:))+ sd*mad(I_mean(:),1)); 

 I_bw2 = bwareaopen(I_bw1,20,4); 

 CC = bwconncomp(I_bw2); 

 I_edge = edge(I_bw2); 

 I_overlay = I_mean; 

 I_overlay(I_edge) = 10*max(I_mean(:))/9;

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

I'm sorry, but it looks like an error has occured. Try removing and re-adding the sweep label. I'll try again in a minute. If this error persists contact [email protected].

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

I'm sorry, but it looks like an error has occured. Try removing and re-adding the sweep label. I'll try again in a minute. If this error persists contact [email protected].

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

I'm sorry, but it looks like an error has occured. Try removing and re-adding the sweep label. I'll try again in a minute. If this error persists contact [email protected].

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

I'm sorry, but it looks like an error has occured. Try removing and re-adding the sweep label. I'll try again in a minute. If this error persists contact [email protected].

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

sweep-ai commented on August 11, 2024

Hey @retroam,

I've started working on turning the app into a CLI tool. The plan is to use the existing run_analysis function as the main entry point for the CLI tool and add argument parsing to specify the necessary parameters. I'll be creating a new file to handle the command line interface and making a small modification to the run_analysis.py file.

Give me a minute!

Some code snippets I looked at (click to expand). If some file is missing from here, you can mention the path in the ticket description.

Calcium-analysis/old/batch_process.py

Lines 1 to 15 in ee83d57

 
 import numpy as np 

 import scipy 

 import matcompat 

 image_fld = 'C:\Users\rka2p\Desktop\Calcium analysis\file_folder' 

 save_fld = 'C:\Users\rka2p\Desktop\Calcium analysis\save_folder' 

 imagestostacks(image_fld,save_fld) 

 #sd from mean of image intensity to set as threshold 

 sd = 7 

 [CC,I_mean,I_bw2,I_overlay] = stackstoroi(save_fld,sd) 

 imagesc(I_overlay) 

 axis off 

 [data,bkg] = roitodata(save_fld)

Calcium-analysis/README.md

Lines 1 to 2 in ee83d57

 # Calcium-analysis 

 Python implementation of calcium imaging analysis pipeline for cardiac myocytes for [Automated image analysis of cardiac myocyte Ca2+ dynamics](https://pubmed.ncbi.nlm.nih.gov/22255377/)

Calcium-analysis/tests/test_image_processing.py

Lines 1 to 19 in ee83d57

 import unittest 

 import numpy as np 

 import os 

 from PIL import Image 

 import image_processing # assuming the module is named image_processing.py 

 class TestImageProcessing(unittest.TestCase): 

 def test_load_image(self): 

 # Test loading a single image 

 image_path = 'path_to_a_test_image.tif' 

 I, info = image_processing.load_image(image_path) 

 self.assertIsInstance(I, np.ndarray) 

 self.assertIsInstance(info, dict) 

 self.assertEqual(I.shape[0], info['ImageLength']) 

 self.assertEqual(I.shape[1], info['ImageWidth']) 

 if __name__ == '__main__':

Calcium-analysis/matlab/imagestostacks.m

Lines 1 to 31 in ee83d57

 function[I_t] = imagestostacks(image_fld,save_fld) 

 % IMAGESTOSTACKS(image_fld,save_fld) assembles tif images in image_fld as 

 % a single matrix which is then saved in save_fld as a .mat file 

 % NOTE: tif files in image_fld have to be labelled as image01...09,10 etc. 

 %Example: 

 % I = imagestostacks(image_fld,save_fld); 

 cd(image_fld); 

 i_bloc = dir('*.tif*'); 

 I_t = []; 

 for j = 1:length(i_bloc) 

 info = imfinfo(i_bloc(j).name); 

 fileName{j} = i_bloc(j).name; 

 num_images = numel(info); 

 I_cum = zeros(info(1,1).Height,info(1,1).Width,num_images-1);%pre-alloc 

 disp(['Loading tif file ' num2str(j) ' of ' num2str(length(i_bloc))]); 

 for k = 1:num_images 

 I = imread(i_bloc(j).name, k, 'Info', info); 

 I_cum(:,:,k) = I; 

 end 

 close all force; 

 I_t = cat(3,I_cum,I_t); 

 clear I_cum; 

 end 

 save([save_fld '\I.mat'],'I_t');

Calcium-analysis/tests/test_image_segmentation.py

Lines 1 to 18 in ee83d57

 import numpy as np 

 import pytest 

 from calcium_analysis.image_segmentation import stack_to_roi 

 def test_stack_to_roi(): 

 # Create a mock image stack 

 image_stack = np.random.rand(100, 100, 10) 

 # Call stack_to_roi with the mock image stack and a standard deviation value 

 labels, I_mean, I_bw2 = stack_to_roi(image_stack, 1) 

 # Assert that the returned labels, mean image, and binary image have the expected shapes and types 

 assert isinstance(labels, np.ndarray) 

 assert labels.shape == (100, 100) 

 assert isinstance(I_mean, np.ndarray) 

 assert I_mean.shape == (100, 100) 

 assert isinstance(I_bw2, np.ndarray) 

 assert I_bw2.shape == (100, 100)

I'm a bot that handles simple bugs and feature requests but I might make mistakes. Please be kind!

from calcium-analysis.

Sweep: Turn app into a cli tool about calcium-analysis HOT 8 CLOSED

Comments (8)

Related Issues (13)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	# Calcium-analysis
	Python implementation of calcium imaging analysis pipeline for cardiac myocytes for [Automated image analysis of cardiac myocyte Ca2+ dynamics](https://pubmed.ncbi.nlm.nih.gov/22255377/)

	import numpy as np
	from skimage import measure
	import os

	def roi_to_data(roi_file, image_file):
	"""
	Extract data from regions of interest in an image stack.

	This function loads a labeled image (from a .npy file), computes
	the mean intensity of each labeled region in each image in a stack,
	and saves the results.
	"""

	# Load the labeled image and the image stack
	labels = np.load(roi_file)
	I_t = np.load(image_file)

	# Extract data from each region in each image
	data = []
	bkg = []
	for I in np.rollaxis(I_t, axis=2):
	# Get properties of labeled regions
	props = measure.regionprops(labels, intensity_image=I)

	# Compute the mean intensity of each region and the background
	avg = [prop.mean_intensity for prop in props]
	I_bkg = I.copy()
	for prop in props:
	I_bkg[prop.coords[:, 0], prop.coords[:, 1]] = np.nan
	bkg_intensity = np.nanmean(I_bkg)

	data.append(avg)
	bkg.append(bkg_intensity)

	# Save the results
	np.save('data.npy', data)
	np.save('bkg.npy', bkg)

	# Return the results

	import unittest
	import numpy as np
	import os
	from PIL import Image
	import image_processing # assuming the module is named image_processing.py

	class TestImageProcessing(unittest.TestCase):

	def test_load_image(self):
	# Test loading a single image
	image_path = 'path_to_a_test_image.tif'
	I, info = image_processing.load_image(image_path)

	self.assertIsInstance(I, np.ndarray)
	self.assertIsInstance(info, dict)
	self.assertEqual(I.shape[0], info['ImageLength'])
	self.assertEqual(I.shape[1], info['ImageWidth'])

	if __name__ == '__main__':

	function[I_t] = imagestostacks(image_fld,save_fld)

	% IMAGESTOSTACKS(image_fld,save_fld) assembles tif images in image_fld as
	% a single matrix which is then saved in save_fld as a .mat file
	% NOTE: tif files in image_fld have to be labelled as image01...09,10 etc.

	%Example:
	% I = imagestostacks(image_fld,save_fld);

	cd(image_fld);
	i_bloc = dir('.tif');
	I_t = [];


	for j = 1:length(i_bloc)
	info = imfinfo(i_bloc(j).name);
	fileName{j} = i_bloc(j).name;
	num_images = numel(info);
	I_cum = zeros(info(1,1).Height,info(1,1).Width,num_images-1);%pre-alloc
	disp(['Loading tif file ' num2str(j) ' of ' num2str(length(i_bloc))]);
	for k = 1:num_images
	I = imread(i_bloc(j).name, k, 'Info', info);
	I_cum(:,:,k) = I;
	end
	close all force;
	I_t = cat(3,I_cum,I_t);
	clear I_cum;
	end

	save([save_fld '\I.mat'],'I_t');


	function[CC,I_mean,I_bw2,I_overlay] = stackstoroi(save_fld,sd)

	% STACKSTOROI(image_fld,sd) loads I.mat from save_fld for image
	% segmentation. sd is how many standard deviations above the mean the
	% threshold is set for segmentation. Outputs:
	% CC: connected components found in I_bw2
	% I_mean: averaged image of I.mat
	% I_bw2: binary image containing image segementation of I.mat
	% I_overlay: overlay of identified ROI over I_mean

	%Example:
	% [CC,I_mean,I_bw2,I_overlay] = stackstoroi(save_fld,sd);

	cd(save_fld);
	m_bloc = dir('I.mat*');
	I_cum = [];
	img_content = [];
	if length(m_bloc)== 11
	for i = 1:11
	I_cum = cat(3,I_cum,importdata(m_bloc(i).name));
	end

	else
	I_cum = importdata(m_bloc(1).name);
	end

	for j = 1:length(I_cum)
	I = I_cum(:,:,j);
	img_content(j) = mean(I(:));
	end
	[marks,locs] = findpeaks(img_content,'minpeakdistance',100);
	I_mean = mean(I_cum(:,:,locs),3);

	%I_bw1 = bwspecial(otsu(I_mean,3));
	I_bw1 = threshold(I_mean,mode(I_mean(:))+ sd*mad(I_mean(:),1));
	I_bw2 = bwareaopen(I_bw1,20,4);
	CC = bwconncomp(I_bw2);
	I_edge = edge(I_bw2);
	I_overlay = I_mean;
	I_overlay(I_edge) = 10*max(I_mean(:))/9;


	import numpy as np
	import scipy
	import matcompat

	image_fld = 'C:\Users\rka2p\Desktop\Calcium analysis\file_folder'
	save_fld = 'C:\Users\rka2p\Desktop\Calcium analysis\save_folder'
	imagestostacks(image_fld,save_fld)

	#sd from mean of image intensity to set as threshold
	sd = 7
	[CC,I_mean,I_bw2,I_overlay] = stackstoroi(save_fld,sd)
	imagesc(I_overlay)
	axis off
	[data,bkg] = roitodata(save_fld)

	import numpy as np
	import pytest
	from calcium_analysis.image_segmentation import stack_to_roi

	def test_stack_to_roi():
	# Create a mock image stack
	image_stack = np.random.rand(100, 100, 10)

	# Call stack_to_roi with the mock image stack and a standard deviation value
	labels, I_mean, I_bw2 = stack_to_roi(image_stack, 1)

	# Assert that the returned labels, mean image, and binary image have the expected shapes and types
	assert isinstance(labels, np.ndarray)
	assert labels.shape == (100, 100)
	assert isinstance(I_mean, np.ndarray)
	assert I_mean.shape == (100, 100)
	assert isinstance(I_bw2, np.ndarray)
	assert I_bw2.shape == (100, 100)