Spaces:
Running on Zero
Running on Zero
File size: 11,331 Bytes
ab1db83 2e66cee ab1db83 7fb7cc3 ab1db83 3188a6e ab1db83 | 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 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 | """CT workspace — paired image+mask generation with anatomy controls."""
from __future__ import annotations
from typing import Any
import gradio as gr
from pipelines import GenerationRequest
from pipelines.ct import generate as generate_ct
from utils.windowing import CT_PRESETS
from viewer.niivue_embed import empty_html, render_viewer
from viewer.colormaps import legend_html
from .presets import CT_ANATOMY_CHOICES, CT_BODY_REGIONS, CT_SAMPLES, XY_CHOICES, Z_CHOICES
def _spacing_default() -> tuple[float, float, float]:
return (1.5, 1.5, 1.5)
def _on_preset(preset_name: str) -> tuple[float, float] | None:
preset = CT_PRESETS.get(preset_name)
if preset is None:
return None
lo, hi = preset.to_min_max()
return lo, hi
def build(spaces_gpu: Any) -> tuple[gr.Group, gr.Button]:
"""Returns the hidden workspace Group and the back-to-home button."""
with gr.Group(visible=False, elem_classes=["workspace"]) as group:
with gr.Row(elem_classes=["workspace-header"]):
back_btn = gr.Button("← Back", elem_classes=["back-btn"], scale=0)
gr.HTML(
'<div class="workspace-title">'
'<span class="ws-dot" style="background:var(--ct);color:var(--ct)"></span>'
'<span class="ws-crumb">NV-Generate</span>'
'<span class="ws-crumb-sep">/</span>'
'<span class="ws-active">CT</span>'
'</div>'
)
gr.HTML(
"""
<div class="ws-intro ws-intro-ct">
<div class="ws-intro-left">
<h2 class="ws-intro-title">NV-Generate · CT</h2>
<p class="ws-intro-desc">
Whole-body synthetic CT volumes with paired 132-class anatomy masks.
Generate balanced training data for segmentation models, augment rare
pathologies with controllable organ and tumor size, or share
privacy-preserving samples for research.
</p>
</div>
<div class="ws-intro-facts">
<div class="ws-fact"><span class="ws-fact-k">Architecture</span><span class="ws-fact-v">MAISI-v2 · Rectified Flow</span></div>
<div class="ws-fact"><span class="ws-fact-k">Body regions</span><span class="ws-fact-v">head · chest · thorax · abdomen · pelvis · lower</span></div>
<div class="ws-fact"><span class="ws-fact-k">Segmentation</span><span class="ws-fact-v">132 classes · paired</span></div>
<div class="ws-fact"><span class="ws-fact-k">Inference</span><span class="ws-fact-v">30 rectified-flow steps</span></div>
<div class="ws-fact"><span class="ws-fact-k">Max volume</span><span class="ws-fact-v">512 × 512 × 768 vox</span></div>
</div>
</div>
"""
)
with gr.Row(elem_classes=["workspace-row"]):
with gr.Column(scale=4, min_width=320, elem_classes=["controls"]):
gr.Markdown("##### Quick presets")
with gr.Row():
sample_btns = [gr.Button(s["label"], size="sm") for s in CT_SAMPLES]
gr.Markdown("##### Conditioning")
generate_masks = gr.Checkbox(
label="Paired anatomy mask · 132 classes",
value=True,
info="Off → image-only generation (faster).",
)
body_region = gr.CheckboxGroup(
choices=CT_BODY_REGIONS,
value=["abdomen"],
label="Body region",
)
anatomy_list = gr.Dropdown(
choices=CT_ANATOMY_CHOICES,
value=["liver"],
multiselect=True,
label="Target anatomies",
)
gr.Markdown("##### Geometry")
dim_xy = gr.Radio(choices=XY_CHOICES, value=256, label="X / Y (voxels)")
dim_z = gr.Radio(choices=Z_CHOICES, value=256, label="Z (voxels)")
with gr.Row(equal_height=True):
sp_x = gr.Slider(1.0, 5.0, value=1.5, step=0.05, label="Spacing X (mm)")
sp_y = gr.Slider(1.0, 5.0, value=1.5, step=0.05, label="Spacing Y (mm)")
sp_z = gr.Slider(0.5, 5.0, value=1.5, step=0.05, label="Spacing Z (mm)")
gr.HTML('<div class="hint">Field of view needs at least 256 mm: increase X voxels or X spacing if shorter.</div>')
gr.Markdown("##### Diffusion")
with gr.Row(equal_height=True):
seed = gr.Number(value=0, label="Seed", precision=0, elem_classes=["seed-field"])
steps = gr.Slider(10, 60, value=30, step=1, label="Inference steps")
generate_btn = gr.Button("Generate volume", variant="primary", elem_classes=["primary-cta"])
status = gr.HTML('<div class="stat-line"><span class="stat-label" style="color:var(--muted)">Idle. Configure parameters and click Generate.</span></div>', elem_classes=["status"])
with gr.Column(scale=8, min_width=520, elem_classes=["viewer-col"]):
gr.HTML(
'<div class="viewer-strip">'
'<span class="viewer-strip-left">Viewport · Multiplanar</span>'
'<span class="viewer-strip-right">Axial · Coronal · Sagittal · 3D</span>'
'</div>'
)
viewer = gr.HTML(empty_html(), elem_classes=["viewer"])
with gr.Row(elem_classes=["preset-row"]):
gr.HTML('<div class="preset-label">Window / level preset</div>')
preset = gr.Radio(
choices=[p.name for p in CT_PRESETS.values()],
value="Soft Tissue",
show_label=False,
container=False,
elem_classes=["preset-radio"],
)
legend = gr.HTML("", elem_classes=["legend-host"], visible=False)
download = gr.File(label="Download generated NIfTI", visible=False, elem_classes=["nv-download"])
# State holding the most recent generation, so the W/L preset radio can
# re-render the viewer without re-running inference.
last_result = gr.State(None)
_PRESET_KEY = {"Soft Tissue": "soft_tissue", "Lung": "lung", "Bone": "bone", "Brain": "brain"}
# ---- handlers ----
def _generate(generate_masks, body_region, anatomy_list, dim_xy, dim_z, sp_x, sp_y, sp_z, seed, steps, preset):
req = GenerationRequest(
model="ct",
output_size=(int(dim_xy), int(dim_xy), int(dim_z)),
spacing=(float(sp_x), float(sp_y), float(sp_z)),
seed=int(seed),
num_steps=int(steps),
body_region=list(body_region) if generate_masks else None,
anatomy_list=list(anatomy_list) if anatomy_list else ["liver"],
generate_masks=bool(generate_masks),
)
try:
result = generate_ct(req)
except Exception as e:
return (
empty_html(f"Generation failed: {e}"),
gr.update(visible=False, value=""),
gr.update(visible=False, value=None),
f'<div class="stat-line"><span class="stat-err">✕ Generation failed</span> <span class="stat-chip"><span class="stat-k">ERR</span><span class="stat-v">{e}</span></span></div>',
None,
)
wm = CT_PRESETS.get(_PRESET_KEY.get(preset, "soft_tissue"))
window_min, window_max = wm.to_min_max() if wm else (None, None)
html = render_viewer(
volume_path=result.volume_path,
mask_path=result.mask_path,
colormap="gray", # CT base is grayscale; W/L preset drives windowing via cal_min/max
used_label_ids=list(result.used_anatomy_labels.keys()),
window_min=window_min,
window_max=window_max,
)
legend_str = legend_html(result.used_anatomy_labels) if result.mask_path else ""
files = [result.volume_path] + ([result.mask_path] if result.mask_path else [])
stat = (
'<div class="stat-line">'
'<span class="stat-mark"></span>'
'<span class="stat-label">Generated</span>'
f'<span class="stat-chip"><span class="stat-k">runtime</span><span class="stat-v">{result.runtime_seconds:.1f}s</span></span>'
f'<span class="stat-chip"><span class="stat-k">seed</span><span class="stat-v">{result.seed}</span></span>'
f'<span class="stat-chip"><span class="stat-k">steps</span><span class="stat-v">{req.num_steps}</span></span>'
f'<span class="stat-chip"><span class="stat-k">size</span><span class="stat-v">{req.output_size[0]}³</span></span>'
'</div>'
)
return (
html,
gr.update(visible=bool(legend_str), value=legend_str),
gr.update(visible=True, value=files),
stat,
result,
)
def _reapply_preset(preset, result):
if not result or not getattr(result, "volume_path", None):
return gr.update()
wm = CT_PRESETS.get(_PRESET_KEY.get(preset, "soft_tissue"))
window_min, window_max = wm.to_min_max() if wm else (None, None)
return render_viewer(
volume_path=result.volume_path,
mask_path=result.mask_path,
colormap="gray",
used_label_ids=list(result.used_anatomy_labels.keys()) if result.used_anatomy_labels else [],
window_min=window_min,
window_max=window_max,
)
decorated = spaces_gpu(_generate) if spaces_gpu else _generate
(
generate_btn.click(
lambda: gr.update(value="Generating volume…", interactive=False),
outputs=[generate_btn],
)
.then(
decorated,
inputs=[generate_masks, body_region, anatomy_list, dim_xy, dim_z, sp_x, sp_y, sp_z, seed, steps, preset],
outputs=[viewer, legend, download, status, last_result],
show_progress="full",
)
.then(
lambda: gr.update(value="Generate volume", interactive=True),
outputs=[generate_btn],
)
)
preset.change(_reapply_preset, inputs=[preset, last_result], outputs=[viewer])
for btn, sample in zip(sample_btns, CT_SAMPLES):
def _apply(s=sample):
return (
s["body_region"],
s["anatomy_list"],
s["xy"], s["z"],
s["spacing"][0], s["spacing"][1], s["spacing"][2],
)
btn.click(_apply, outputs=[body_region, anatomy_list, dim_xy, dim_z, sp_x, sp_y, sp_z])
return group, back_btn
|