-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathevaluate.html
More file actions
338 lines (298 loc) · 12.7 KB
/
evaluate.html
File metadata and controls
338 lines (298 loc) · 12.7 KB
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
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>SegRap2025 Challenge</title>
<link rel="stylesheet" href="shared.css">
<style>
.evaluate-section {
margin-bottom: 60px;
}
.evaluate-content {
background: white;
border-radius: 12px;
padding: 10px;
margin-bottom: 0px;
}
.leaderboard-table {
width: 100%;
border-collapse: collapse;
margin: 20px 0;
}
.leaderboard-table th,
.leaderboard-table td {
padding: 12px;
text-align: left;
border-bottom: 1px solid #eee;
}
.leaderboard-table th {
font-weight: 500;
color: var(--text-primary);
background: #f8f9fa;
}
.leaderboard-table td {
color: var(--text-secondary);
}
.rank {
font-weight: 600;
color: var(--primary-color);
}
.team-name {
color: var(--text-primary);
font-weight: 500;
}
.metric-value {
font-family: monospace;
}
.evaluation-metrics {
margin-top: 20px;
}
.metric-item {
margin-bottom: 20px;
}
.metric-name {
font-weight: 500;
color: var(--text-primary);
font-size: 20px;
margin-bottom: 8px;
margin-top: 0px;
}
.metric-name-sub {
font-weight: 500;
color: var(--text-secondary);
font-size: 16px;
margin-bottom: 8px;
margin-top: 0px;
}
.metric-description {
color: var(--text-secondary);
font-size: 0.95em;
line-height: 1.5;
width: 100%;
}
.metric-description ul {
list-style: none;
padding: 0;
margin: 0;
width: 100%;
}
.metric-description li {
margin-bottom: 8px;
}
.submit-section {
margin-top: 30px;
width: 100%;
}
.submit-button {
display: inline-block;
padding: 12px 24px;
background: var(--primary-color);
color: white;
border-radius: 4px;
text-decoration: none;
transition: background 0.2s ease;
}
.submit-button:hover {
background: var(--primary-dark);
}
.tree::before {
color: rgb(0, 0, 0);
content: "Example:";
font-weight: bold;
display: block;
margin-bottom: 5px;
}
.tree ul {
color: rgb(0, 0, 0);
font-size: 0.95em;
list-style-type: none;
margin-left: 10px;
padding-left: 0;
}
.tree li::before {
content: "├── ";
font-size: 0.9em;
margin-left: 10px;
color: rgb(0, 0, 0);
}
.news-list {
list-style: none;
padding: 0;
margin: 0;
}
.news-list li {
position: relative;
padding-left: 25px;
margin-top: 1%;
/* margin-bottom: 20px; */
line-height: 1.3;
color: #000000;
font-size: 0.95em;
font-weight: 400;
}
.news-list li::before {
content: "•";
position: absolute;
left: 10px;
color: #00bfa5;
}
/* .news-list a {
color: #00bfa5;
text-decoration: none;
font-weight: 500;
transition: color 0.2s ease;
} */
.news-list a:hover {
color: #00897b;
}
/* .tree li:last-child::before {
content: "└── ";
color: rgb(0, 0, 0);
} */
</style>
</head>
<body>
<div class="site-header">
<div class="header-image" style="text-align: center;">
<img src="logo_web.png" alt="SegRap Logo" style="width: 50%; height: auto;">
</div>
<div class="header-container">
<div class="content-section">
<nav class="nav-links">
<a href="index.html">Home</a>
<a href="tasks.html">Tasks</a>
<a href="dataset.html">Dataset</a>
<a href="evaluate.html" class="active">Evaluate</a>
<a href="prizes.html">Prizes</a>
<a href="leaderboard.html">Leaderboard</a>
<a href="organizing.html">Organizing</a>
<a href="contact.html">Contact</a>
</nav>
</div>
</div>
</div>
<main class="main-content">
<!-- Leaderboards Section -->
<section id="leaderboards" class="evaluate-section">
<h1 class="section-title">Evaluate and Ranking</h1>
<!-- Segmentation Leaderboard -->
<div class="evaluate-content">
<div class="metric-name">1. Two-phase Challenge Submission</div>
<div class="metric-name-sub"><strong>1) Validation Phase</strong></div>
<p class="metric-description">
For the first phase (validation phase), the participants are required to submit the output of their
algorithms as a single compressed zip file to the organization team (<em>segrap2025@163.com</em>).
The submitted zip files should be formatted like the one below. Make sure the results in the
submitted zip file all be matched with the validation cases one-to-one (2 classes for Task01 and 6
classes for Task02). Otherwise, the results are considered invalid submissions and no score will be
generated.
</p>
<div class="tree">
<ul> team_name/
<ul> Task0*_results/
<ul>
<li>segrap_0001.nii.gz</li>
<li>......</li>
<li>segrap_xxxx.nii.gz</li>
</ul>
</ul>
</ul>
</div>
<br>
<p>
<em>Note</em>: A maximum of 5 submissions are allowed during the validation stage to ensure fair
evaluation and prevent overfitting.
</p>
<br>
<div class="metric-name-sub"><strong>2) Testing Phase</strong></div>
<p class="metric-description">
For the second phase (test phase), participants must prepare and submit their <strong>Docker
containers</strong> along with a <strong>short paper</strong> outlining their method before 31st
August 2025.
<p>
<br>
<p class="metric-description"> <em><strong>Docker container:</strong></em></p>
<ul class="news-list">
<li>The Docker submission tutorial can be found <a
href="https://github.com/HiLab-git/SegRap2025_Docker">here</a>.</li>
<li>Memory constraint: GPU memory usage less than 24 GB, CPU memory usage less than 64 GB</li>
<li>Execution time constraint: no more than 3 minutes per case</li>
<li>Once you successfully build your Docker container, save it to a zipped file
'<span style="color: red;">Task_{Task_id}_{TeamName}.zip</span>'
and upload to cloud
platform such as <a href="https://workspace.google.com/products/drive/">Google Drive</a> and <a
href="https://wangpan.baidu.com/">BaiduNetDisk</a>. To submit your algorithm we ask you to
send us the download link and step-by-step command to <strong>segrap2025@163.com</strong> with
the Subject '<span style="color: red;">SegRap2025 Testing Phase Submission - Task_id TeamName
Docker Container</span>'.
</li>
</ul>
<br>
<p class="metric-description"><em><strong>Short paper:</strong></em> please provide a description of
your model highlighting the main features using Template <a
href="https://conferences.miccai.org/2025/files/downloads/MICCAI2025-LaTeX-Template.zip">Latex</a>
or <a href="https://conferences.miccai.org/2025/files/downloads/MICCAI2025-Word-Template.zip">MS
Word</a>. Basic information and the description of the model must include the following
details:
</p>
<ul class="news-list">
<li>Team name, Team members (maximum three) (names, emails, affiliations)</li>
<li>Framework (ie, MONAI, nnUNet, etc.)</li>
<li>Model architecture</li>
<li>Number of layers</li>
<li>Convolution kernel size</li>
<li>Initialization</li>
<li>Optimizer</li>
<li>Cross-validation used?</li>
<li>Number of epochs</li>
<li>Number of trainable parameters</li>
<li>Learning Rate and schedule</li>
<li>Loss Function</li>
<li>Dimensionality of input/output (ie, 2D, 3D, 2.5D, etc.)</li>
<li>Batch Size</li>
<li>Preprocessing steps used (data normalization, creation of patches, etc.)</li>
<li>Data Augmentation steps (rotation, flipping, scaling, blur, noise, etc.)</li>
<li>Pretrained model used? (allowed, but it needs to be publicly available)</li>
<li>Number of models trained for final submission</li>
<li>Post-Processing Steps (ensemble network, voting, label fusion, etc.)</li>
<li>Clearly state which aspects are original work (if any) or already existing work</li>
<li>Include relevant citations, as well as if existing code/software libraries/packages were used
</li>
<li>Which cases were included in the training and testing (all cases, only labeled cases, only
paired cases, etc.)</li>
<li>Training/validation/testing data splits</li>
<li>Hyperparameter tuning performed</li>
<li>Training time</li>
</ul>
</div>
<!-- Evaluation metrics -->
<div class="evaluate-content">
<div class="metric-name">2. Evaluation metrics</div>
<p class="metric-description">
Two classical medical segmentation metrics: <strong>Dice Similarity Coefficient (DSC)</strong>, and
<strong>normalized surface dice (NSD)</strong>, will be used to assess different aspects of the
performance of the segmentation methods.
</p>
</div>
<!-- Ranking -->
<div class="evaluate-content">
<div class="metric-name">3. Ranking</div>
<p class="metric-description">
Firstly, for each cohort (set), we calculate the average DSC and NSD across all the patients for
each class, respectively. Secondly, each participant will be ranked based on the class-level DSC and
NSD, with 2 × 2 or 6 × 2 rankings. Then, rankings for all classes are averaged in each
cohort (set). Finally, rankings for all cohorts (sets) are averaged and then normalized by the
number of teams. At the same time, we will take the statistical ranking. (Allow equal teams if there
is no significant difference).</p>
<br>
<p class="metric-description">
In addition, if the submissions have some missing results on test cases, the corresponding class's
DSC and NSD will be set to 0 and 0 for ranking. For example, a test case missed a class and the
ranking value of this class in average DSC and NSD will degrade.
</p>
</div>
</section>
</main>
</body>
</html>