Python based (no gui) way of estimating diameters

[mat10d]

I would like to replicate the diameter calibration outside of the GUI, for the pretrained models provided. I am running:

def estimate_cell_diameters(data, channels=[2,3], gpu=False):
    """Estimate diameter directly using Cellpose model's size estimator"""
    model = Cellpose(model_type="cyto3", gpu=gpu)
    diams, _ = model.sz.eval(data, channels=channels)
    return np.maximum(5.0, diams)

This is based on the calibrate size function in the GUI, found here. Is this a reliable way of getting optimal diameters in a purely python way?

Furthermore, every time I run this function, I get different calibrated diameters. Is this expected?

[carsen-stringer]

that is correct and should be deterministic, can you please post what is written from the cellpose logger?

[mat10d]

Sorry for not updating--this works! I think I was saving the diameters incorrectly.

def estimate_diameters(
    data,
    dapi_index,
    cyto_index,
    channels=[2, 3],  # Default channels for cell estimation
    cyto_model="cyto3",
    cellpose_kwargs=dict(flow_threshold=0.4, cellprob_threshold=0),
    gpu=False,
    logscale=True,
):
    """Estimate optimal cell diameter using Cellpose's diameter estimation.

    Args:
        data (numpy.ndarray): Multichannel image data
        dapi_index (int): Index of DAPI channel
        cyto_index (int): Index of cytoplasmic channel
        channels (list): Channel indices for diameter estimation [cytoplasm, nuclei]
        cyto_model (str): Cellpose model type to use
        cellpose_kwargs (dict): Additional keyword arguments for Cellpose
        gpu (bool): Whether to use GPU
        logscale (bool): Whether to apply log scaling to image

    Returns:
        tuple: Estimated diameters for (nuclei, cells)
    """
    # Prepare RGB image
    log_kwargs = cellpose_kwargs.pop(
        "log_kwargs", dict()
    )  # Extract log_kwargs from cellpose_kwargs
    rgb = prepare_cellpose(
        data, dapi_index, cyto_index, logscale, log_kwargs=log_kwargs
    )

    # Find optimal nuclei diameter
    print("Estimating nuclei diameters...")
    model_nuclei = Cellpose(model_type="nuclei", gpu=gpu)
    diam_nuclear, _ = model_nuclei.sz.eval(rgb, channels=[3, 0])
    diam_nuclear = np.maximum(5.0, diam_nuclear)
    diam_nuclear = float(diam_nuclear)
    print(f"Estimated nuclear diameter: {diam_nuclear:.1f} pixels")

    # Find optimal cell diameter
    print("Estimating cell diameters...")
    model_cyto = Cellpose(model_type=cyto_model, gpu=gpu)
    diam_cell, _ = model_cyto.sz.eval(rgb, channels=channels)
    diam_cell = np.maximum(5.0, diam_cell)
    diam_cell = float(diam_cell)
    print(f"Estimated cell diameter: {diam_cell:.1f} pixels")

    return diam_nuclear, diam_cell

[carsen-stringer]

phew thanks for letting me know!