All Downloads are FREE. Search and download functionalities are using the official Maven repository.

org.scijava.ops.image.coloc.saca.WtKendallTau Maven / Gradle / Ivy

The newest version!
/*-
 * #%L
 * Image processing operations for SciJava Ops.
 * %%
 * Copyright (C) 2014 - 2024 SciJava developers.
 * %%
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 * 
 * 1. Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 * 
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 * #L%
 */

package org.scijava.ops.image.coloc.saca;

import java.lang.Double;
import java.util.Arrays;
import java.util.Comparator;
import java.util.Random;

import org.scijava.ops.image.coloc.IntComparator;

/**
 * Helper class for Spatially Adaptive Colocalization Analysis (SACA) framework.
 * This class is used by the AdaptiveSmoothedKendallTau class.
 *
 * @author Shulei Wang
 * @author Ellen TA Dobson
 * @author Curtis Rueden
 */

final class WtKendallTau {

	private static final Comparator SORT_X = (row1, row2) -> Double
		.compare(row1[0], row2[0]);

	private static final Comparator SORT_Y = (row1, row2) -> Double
		.compare(row1[1], row2[1]);

	public static double calculate(final double[] X, final double[] Y,
		final double[] W, final double[][] combinedData, final int[] rankedindex,
		final double[] rankedw, final int[] index1, final int[] index2,
		final double[] w1, final double[] w2, final double[] cumw, final Random rng)
	{

		final double[][] rankedData = rank(X, Y, W, combinedData, rng);

		for (int i = 0; i < X.length; i++) {
			rankedindex[i] = (int) rankedData[i][0];
			rankedw[i] = rankedData[i][2];
		}

		final double swap = sort(rankedindex, rankedw, Integer::compare, index1,
			index2, w1, w2, cumw);
		final double tw = totw(W) / 2;

		// compute tau and check for NaN results
		double tauTemp = (tw - 2 * swap) / tw;
		if (Double.isNaN(tauTemp)) {
			tauTemp = 0.0;
		}

		final double tau = tauTemp;

		return tau;
	}

	private static double totw(final double[] w) {
		double sumw = 0;
		double sumsquarew = 0;

		for (int i = 0; i < w.length; i++) {
			sumw += w[i];
			sumsquarew += w[i] * w[i];
		}

		final double result = sumw * sumw - sumsquarew;

		return result;
	}

	private static double[][] rank(final double[] IX, final double[] IY,
		final double[] IW, final double[][] combinedData, final Random rng)
	{

		for (int i = 0; i < IX.length; i++) {
			combinedData[i][0] = IX[i];
			combinedData[i][1] = IY[i];
			combinedData[i][2] = IW[i];
		}

		// sort and rank X
		Arrays.sort(combinedData, SORT_X);
		rank1D(combinedData, 0, rng);

		// sort and rank Y
		Arrays.sort(combinedData, SORT_Y);
		rank1D(combinedData, 1, rng);

		return combinedData;
	}

	private static void rank1D(final double[][] combinedData, final int dim,
		final Random rng)
	{
		int start = 0;
		int end = 1;
		int rank = 1;

		while (start < combinedData.length) {
			if (end < combinedData.length &&
				combinedData[start][dim] == combinedData[end][dim])
			{
				// tied value; count how many tied values there are in a row
				do {
					end++;
				}
				while (end < combinedData.length &&
					combinedData[start][dim] == combinedData[end][dim]);
				// now assign unique rank randomly over these indices -- Fisher-Yates
				// shuffle!
				for (int i = start; i < end - 1; i++) {
					final int newIndex = start + rng.nextInt(end - start);
					final double[] tmp = combinedData[i];
					combinedData[i] = combinedData[newIndex];
					combinedData[newIndex] = tmp;
				}
				rank += end - start;
			}
			for (int i = start; i < end; i++) {
				combinedData[i][dim] = rank++;
			}
			start = end;
			end++;
		}
	}

	private static double sort(final int[] data, final double[] weight,
		final IntComparator comparator, final int[] index1, final int[] index2,
		final double[] w1, final double[] w2, final double[] cumw)
	{
		double swap = 0;
		double tempswap;
		final int n = data.length;
		int step = 1;
		int begin;
		int begin2;
		int end;
		int k;

		for (int i = 0; i < n; i++) {
			index1[i] = data[i];
			w1[i] = weight[i];
		}

		while (step < n) {
			begin = 0;
			k = 0;
			cumw[0] = w1[0];
			for (int i = 1; i < n; i++) {
				cumw[i] = cumw[i - 1] + w1[i];
			}

			while (true) {
				begin2 = begin + step;
				end = begin2 + step;
				if (end > n) {
					if (begin2 > n) break;
					end = n;
				}
				int i = begin;
				int j = begin2;
				while (i < begin2 && j < end) {
					if (comparator.compare(index1[i], index1[j]) > 0) {
						if (i == 0) {
							tempswap = w1[j] * cumw[begin2 - 1];
						}
						else {
							tempswap = w1[j] * (cumw[begin2 - 1] - cumw[i - 1]);
						}
						swap = swap + tempswap;
						index2[k] = index1[j];
						w2[k++] = w1[j++];
					}
					else {
						index2[k] = index1[i];
						w2[k++] = w1[i++];
					}
				}
				if (i < begin2) {
					while (i < begin2) {
						index2[k] = index1[i];
						w2[k++] = w1[i++];
					}
				}
				else {
					while (j < end) {
						index2[k] = index1[j];
						w2[k++] = w1[j++];
					}
				}
				begin = end;
			}
			if (k < n) {
				while (k < n) {
					index2[k] = index1[k];
					w2[k] = w1[k];
					k++;
				}
			}
			for (int i = 0; i < n; i++) {
				index1[i] = index2[i];
				w1[i] = w2[i];
			}

			step *= 2;
		}
		return swap;
	}
}




© 2015 - 2025 Weber Informatics LLC | Privacy Policy