utils/multi_task_sliding_window_inference.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri Apr  1 19:18:58 2022

@author: jma
"""

from typing import Any, Callable, List, Sequence, Tuple, Union
import torch
import torch.nn.functional as F
from monai.data.utils import compute_importance_map, dense_patch_slices, get_valid_patch_size
from monai.utils import BlendMode, PytorchPadMode, fall_back_tuple, look_up_option


__all__ = ["multi_task_sliding_window_inference"]

def multi_task_sliding_window_inference(
    inputs: torch.Tensor,
    roi_size: Union[Sequence[int], int],
    sw_batch_size: int,
    predictor: Callable[..., torch.Tensor],
    overlap = 0.25,
    mode = "constant",
    sigma_scale = 0.125,
    padding_mode = "constant",
    cval = 0.0,
    sw_device = None,
    device = None,
    *args: Any,
    **kwargs: Any,
) -> torch.Tensor:
    """
    Sliding window inference on `inputs` with `predictor`.

    When roi_size is larger than the inputs' spatial size, the input image are padded during inference.
    To maintain the same spatial sizes, the output image will be cropped to the original input size.

    Args:
        inputs: input image to be processed (assuming NCHW[D])
        roi_size: the spatial window size for inferences.
            When its components have None or non-positives, the corresponding inputs dimension will be used.
            if the components of the `roi_size` are non-positive values, the transform will use the
            corresponding components of img size. For example, `roi_size=(32, -1)` will be adapted
            to `(32, 64)` if the second spatial dimension size of img is `64`.
        sw_batch_size: the batch size to run window slices.
        predictor: given input tensor `patch_data` in shape NCHW[D], `predictor(patch_data)`
            should return a prediction with the same spatial shape and batch_size, i.e. NMHW[D];
            where HW[D] represents the patch spatial size, M is the number of output channels, N is `sw_batch_size`.
        overlap: Amount of overlap between scans.
        mode: {``"constant"``, ``"gaussian"``}
            How to blend output of overlapping windows. Defaults to ``"constant"``.

            - ``"constant``": gives equal weight to all predictions.
            - ``"gaussian``": gives less weight to predictions on edges of windows.

        sigma_scale: the standard deviation coefficient of the Gaussian window when `mode` is ``"gaussian"``.
            Default: 0.125. Actual window sigma is ``sigma_scale`` * ``dim_size``.
            When sigma_scale is a sequence of floats, the values denote sigma_scale at the corresponding
            spatial dimensions.
        padding_mode: {``"constant"``, ``"reflect"``, ``"replicate"``, ``"circular"``}
            Padding mode for ``inputs``, when ``roi_size`` is larger than inputs. Defaults to ``"constant"``
            See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.pad.html
        cval: fill value for 'constant' padding mode. Default: 0
        sw_device: device for the window data.
            By default the device (and accordingly the memory) of the `inputs` is used.
            Normally `sw_device` should be consistent with the device where `predictor` is defined.
        device: device for the stitched output prediction.
            By default the device (and accordingly the memory) of the `inputs` is used. If for example
            set to device=torch.device('cpu') the gpu memory consumption is less and independent of the
            `inputs` and `roi_size`. Output is on the `device`.
        args: optional args to be passed to ``predictor``.
        kwargs: optional keyword args to be passed to ``predictor``.

    Note:
        - input must be channel-first and have a batch dim, supports N-D sliding window.

    """
    num_spatial_dims = len(inputs.shape) - 2
    if overlap < 0 or overlap >= 1:
        raise AssertionError("overlap must be >= 0 and < 1.")

    # determine image spatial size and batch size
    # Note: all input images must have the same image size and batch size
    image_size_ = list(inputs.shape[2:])
    batch_size = inputs.shape[0]

    if device is None:
        device = inputs.device
    if sw_device is None:
        sw_device = inputs.device

    roi_size = fall_back_tuple(roi_size, image_size_)
    # in case that image size is smaller than roi size
    image_size = tuple(max(image_size_[i], roi_size[i]) for i in range(num_spatial_dims))
    pad_size = []
    for k in range(len(inputs.shape) - 1, 1, -1):
        diff = max(roi_size[k - 2] - inputs.shape[k], 0)
        half = diff // 2
        pad_size.extend([half, diff - half])
    inputs = F.pad(inputs, pad=pad_size, mode=mode, value=cval)

    scan_interval = _get_scan_interval(image_size, roi_size, num_spatial_dims, overlap)

    # Store all slices in list
    slices = dense_patch_slices(image_size, roi_size, scan_interval)
    num_win = len(slices)  # number of windows per image
    total_slices = num_win * batch_size  # total number of windows

    # Create window-level importance map
    importance_map = compute_importance_map(
        get_valid_patch_size(image_size, roi_size), mode="gaussian", sigma_scale=sigma_scale, device=device
    )

    # Perform predictions
    output_image, count_map = torch.tensor(0.0, device=device), torch.tensor(0.0, device=device)
    output_dist = torch.tensor(0.0, device=device)
    _initialized = False
    for slice_g in range(0, total_slices, sw_batch_size):
        slice_range = range(slice_g, min(slice_g + sw_batch_size, total_slices))
        unravel_slice = [
            [slice(int(idx / num_win), int(idx / num_win) + 1), slice(None)] + list(slices[idx % num_win])
            for idx in slice_range
        ]
        window_data = torch.cat([inputs[win_slice] for win_slice in unravel_slice]).to(sw_device)
        seg_logit, seg_dist = predictor(window_data)# .to(device)  # batched patch segmentation
        seg_logit = torch.nn.functional.interpolate(seg_logit, size=roi_size, mode="bilinear", align_corners=False)
        seg_logit = torch.softmax(seg_logit, dim=1)
        seg_dist = torch.nn.functional.interpolate(seg_dist, size=roi_size, mode="bilinear", align_corners=False) 
        seg_dist = torch.sigmoid(seg_dist)

        if not _initialized:  # init. buffer at the first iteration
            output_classes = seg_logit.shape[1]
            dist_class = seg_dist.shape[1]
            output_shape = [batch_size, output_classes] + list(image_size)
            output_dist_shape = [batch_size, dist_class] + list(image_size)
            # allocate memory to store the full output and the count for overlapping parts
            output_image = torch.zeros(output_shape, dtype=torch.float32, device=device)
            output_dist = torch.zeros(output_dist_shape, dtype=torch.float32, device=device)
            count_map = torch.zeros(output_shape, dtype=torch.float32, device=device)
            count_dist_map = torch.zeros(output_dist_shape, dtype=torch.float32, device=device)
            _initialized = True

        # store the result in the proper location of the full output. Apply weights from importance map.
        for idx, original_idx in zip(slice_range, unravel_slice):
            output_image[original_idx] += importance_map * seg_logit[idx - slice_g]
            output_dist[original_idx] += importance_map * seg_dist[idx - slice_g]
            count_map[original_idx] += importance_map
            count_dist_map[original_idx] += importance_map

    # account for any overlapping sections
    output_image = output_image / count_map
    output_dist = output_dist / count_dist_map

    final_slicing: List[slice] = []
    for sp in range(num_spatial_dims):
        slice_dim = slice(pad_size[sp * 2], image_size_[num_spatial_dims - sp - 1] + pad_size[sp * 2])
        final_slicing.insert(0, slice_dim)
    while len(final_slicing) < len(output_image.shape):
        final_slicing.insert(0, slice(None))
    return output_image[final_slicing], output_dist[final_slicing]



def _get_scan_interval(
    image_size: Sequence[int], roi_size: Sequence[int], num_spatial_dims: int, overlap: float
) -> Tuple[int, ...]:
    """
    Compute scan interval according to the image size, roi size and overlap.
    Scan interval will be `int((1 - overlap) * roi_size)`, if interval is 0,
    use 1 instead to make sure sliding window works.

    """
    if len(image_size) != num_spatial_dims:
        raise ValueError("image coord different from spatial dims.")
    if len(roi_size) != num_spatial_dims:
        raise ValueError("roi coord different from spatial dims.")

    scan_interval = []
    for i in range(num_spatial_dims):
        if roi_size[i] == image_size[i]:
            scan_interval.append(int(roi_size[i]))
        else:
            interval = int(roi_size[i] * (1 - overlap))
            scan_interval.append(interval if interval > 0 else 1)
    return tuple(scan_interval)