imgal 0.3.0

A fast and open-source scientific image processing and algorithm library.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78

use std::collections::HashMap;

use ndarray::{Array2, ArrayBase, AsArray, Axis, Dimension, IxDyn, ViewRepr};
use rayon::prelude::*;

use crate::prelude::*;
use crate::statistics::pearson;

/// Compute the Pearson correlation coefficient between two n-dimensional images
/// and a ROI map.
///
/// # Description
///
/// Computes the Pearson correlation coefficient, a measure of linear
/// correlation between two sets of n-dimensional images and a ROI map. This
/// function iterates through each ROI in the map and computes the correlation
/// coefficient. Returning a `HashMap` of Pearson correlation coefficient values
/// and ROI label IDs.
///
/// # Arguments
///
/// * `data_a`: The first n-dimensional image for Pearson colocalization
///   analysis.
/// * `data_b`: The second n-dimensional image for Pearson colocalization
///   analysis.
/// * `rois`: A map of point clouds representing Regions of Interest (ROIs).
///   The individual ROIs must have the same dimensionality as the input data.
/// * `threads`: The requested number of threads to use for parallel execution.
///   If `None` or `Some(1)` sequential execution is used. If `Some(0)`, then
///   the maximum available parallelism is used. Thread counts are clamped to
///   the systems maximum.
///
/// # Returns
///
/// * `Ok(HashMap<u64, f64>)`: A `HashMap` where the keys are the ROI label IDs
///   and values are the Pearson correlation coefficients for each ROI
///   respectively.
/// * `Err(ImgalError)`: If `data_a.len() != data_b.len()`. If `data_a.len()` or
///   `data_b.len()` is <= 2.
#[inline]
pub fn pearson_roi_coloc<'a, T, A, D>(
    data_a: A,
    data_b: A,
    rois: &HashMap<u64, Array2<usize>>,
    threads: Option<usize>,
) -> Result<HashMap<u64, f64>, ImgalError>
where
    A: AsArray<'a, T, D>,
    D: Dimension,
    T: 'a + AsNumeric,
{
    let data_a: ArrayBase<ViewRepr<&'a T>, IxDyn> = data_a.into().into_dyn();
    let data_b: ArrayBase<ViewRepr<&'a T>, IxDyn> = data_b.into().into_dyn();
    let per_roi_pearson_corr = |k: u64, v: &Array2<usize>| -> Result<(u64, f64), ImgalError> {
        let n = v.dim().0;
        let mut buf_a: Vec<T> = Vec::with_capacity(n);
        let mut buf_b: Vec<T> = Vec::with_capacity(n);
        let roi_coords = v.lanes(Axis(1));
        roi_coords.into_iter().for_each(|p| {
            let pos_buf;
            let pos = if let Some(coord) = p.as_slice() {
                coord
            } else {
                pos_buf = p.to_vec();
                pos_buf.as_slice()
            };
            buf_a.push(data_a[IxDyn(pos)]);
            buf_b.push(data_b[IxDyn(pos)]);
        });
        let corr = pearson(&buf_a, &buf_b, None)?;
        Ok((k, corr))
    };
    par!(threads,
        seq_exp: rois.iter().map(|(&k, v)| per_roi_pearson_corr(k, v))
            .collect::<Result<HashMap<u64, f64>, ImgalError>>(),
        par_exp: rois.into_par_iter().map(|(&k, v)| per_roi_pearson_corr(k, v))
            .collect::<Result<HashMap<u64, f64>, ImgalError>>())
}