-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathv0.html
349 lines (314 loc) · 18 KB
/
v0.html
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
339
340
341
342
343
344
345
346
347
348
349
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>MedMNIST</title>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="keywords" content="MedMNIST, classification, decathlon, AutoML, medical image analysis">
<meta name="description" content="MedMNIST Project Page">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" type="text/css" href="assets/v1/project.css" media="screen">
<link href="assets/v1/favicon.ico" rel="icon" type="image/x-icon" />
</head>
<body>
<div id="content">
<div id="content-inner">
<div class="section head">
<h1>MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis</h1>
<br>
<div class="authors">
<a href="https://jiancheng-yang.com/" target="_blank">Jiancheng Yang</a>
Rui Shi
<a href="https://scholar.google.com/citations?user=eUbmKwYAAAAJ" target="_blank">Bingbing Ni</a>
<a href="https://scholar.google.com/citations?user=2cX5y8kAAAAJ" target="_blank">Bilian Ke</a>
</div>
<div class="affiliations">
Shanghai Jiao Tong University, Shanghai, China
</div>
</div>
<center>
<font size="3">
Paper
[<a href="https://arxiv.org/abs/2010.14925" target="_blank">ISBI'21</a>]
Code
[<a href="https://github.com/MedMNIST/MedMNIST" target="_blank">Github</a>]
Dataset
[<a href="https://doi.org/10.5281/zenodo.4269852" target="_blank">Zenodo</a>]
</font>
</center>
<center><img src="assets/v1/imgs/overview.jpg" border="0" width="80%"></center>
<div class="section" id="abstract">
<h2>Abstract</h2>
<p style="text-align:justify; text-justify:inter-ideograph">
We present MedMNIST, a collection of 10 pre-processed medical open datasets. MedMNIST is
standardized to perform classification tasks on lightweight 28 * 28 images, which requires no
background knowledge. Covering the primary data modalities in medical image analysis, it is diverse
on data scale (from 100 to 100,000) and tasks (binary/multi-class, ordinal regression and
multi-label). MedMNIST could be used for educational purpose, rapid prototyping, multi-modal machine
learning or AutoML in medical image analysis. Moreover, MedMNIST Classification Decathlon is
designed to benchmark AutoML algorithms on all 10 datasets; We have compared several baseline
methods, including open-source or commercial AutoML tools.
</p>
</div>
<div class="section" id="highlights">
<h2>Key Features</h2>
<ul>
<li style="text-align:justify; text-justify:inter-ideograph">
<b>Educational</b>: Our multi-modal data, from multiple open medical image datasets with
Creative Commons (CC) Licenses, is easy to use for educational purpose.
</li>
<li style="text-align:justify; text-justify:inter-ideograph">
<b>Standardized</b>: Data is pre-processed into same format, which requires no background
knowledge for users.
</li>
<li style="text-align:justify; text-justify:inter-ideograph">
<b>Diverse</b>: The multi-modal datasets covers diverse data scales (from 100 to 100,000) and
tasks (binary/multiclass, ordinal regression and multi-label).
</li>
<li style="text-align:justify; text-justify:inter-ideograph">
<b>Lightweight</b>: The small size of 28 × 28 is friendly for rapid prototyping and
experimenting multi-modal machine learning and AutoML algorithms.
</li>
<p style="text-align:justify; text-justify:inter-ideograph">
Please note that this dataset is <b>NOT</b> intended for clinical use.
</p>
</ul>
</div>
<div class="section" id="dataset">
<h2>Download</h2>
<p style="text-align:justify; text-justify:inter-ideograph">
Please download the dataset(s) via <b><a href="https://doi.org/10.5281/zenodo.4269852" target="_blank">Zenodo</a></b>. You could also use our <a href="https://github.com/MedMNIST/MedMNIST" target="_blank">code</a> to download automatically.
</p>
</div>
<div class="section" id="materials">
<h2>Materials</h2>
<table align="center" , class="tg">
<Caption>An Overview of MedMNIST Dataset</Caption>
<thead>
<tr>
<th class="tg-c3ow">Name</th>
<th class="tg-c3ow">Data Modality</th>
<th class="tg-c3ow">Tasks (# Classes/Labels)</th>
<th class="tg-c3ow"># Training</th>
<th class="tg-c3ow"># Validation</th>
<th class="tg-c3ow"># Test</th>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-c3ow">PathMNIST</td>
<td class="tg-c3ow">Pathology</td>
<td class="tg-c3ow">Multi-Class (9)</td>
<td class="tg-c3ow">89,996</td>
<td class="tg-c3ow">10,004</td>
<td class="tg-c3ow">7,180</td>
</tr>
<tr>
<td class="tg-c3ow">ChestMNIST</td>
<td class="tg-c3ow">Chest X-ray</td>
<td class="tg-c3ow">Multi-Label (14) Binary-Class (2)</td>
<td class="tg-c3ow">78,468</td>
<td class="tg-c3ow">11,219</td>
<td class="tg-c3ow">22,433</td>
</tr>
<tr>
<td class="tg-c3ow">DermaMNIST</td>
<td class="tg-c3ow">Dermatoscope</td>
<td class="tg-c3ow">Multi-Class (7)</td>
<td class="tg-c3ow">7,007</td>
<td class="tg-c3ow">1,003</td>
<td class="tg-c3ow">2,005</td>
</tr>
<tr>
<td class="tg-c3ow">OCTMNIST</td>
<td class="tg-c3ow">OCT</td>
<td class="tg-c3ow">Multi-Class (4)</td>
<td class="tg-c3ow">97,477</td>
<td class="tg-c3ow">10,832</td>
<td class="tg-c3ow">1,000</td>
</tr>
<tr>
<td class="tg-c3ow">PneumoniaMNIST</td>
<td class="tg-c3ow">Chest X-ray</td>
<td class="tg-c3ow">Binary-Class (2)</td>
<td class="tg-c3ow">4,708</td>
<td class="tg-c3ow">524</td>
<td class="tg-c3ow">624</td>
</tr>
<tr>
<td class="tg-c3ow">RetinaMNIST</td>
<td class="tg-c3ow">Fundus Camera</td>
<td class="tg-c3ow">Ordinal Regression (5)</td>
<td class="tg-c3ow">1,080</td>
<td class="tg-c3ow">120</td>
<td class="tg-c3ow">400</td>
</tr>
<tr>
<td class="tg-c3ow">BreastMNIST</td>
<td class="tg-c3ow">Breast Ultrasound</td>
<td class="tg-c3ow">Binary-Class (2)</td>
<td class="tg-c3ow">546</td>
<td class="tg-c3ow">78</td>
<td class="tg-c3ow">156</td>
</tr>
<tr>
<td class="tg-c3ow">OrganMNIST_Axial</td>
<td class="tg-c3ow">Abdominal CT</td>
<td class="tg-c3ow">Multi-Class (11)</td>
<td class="tg-c3ow">34,581</td>
<td class="tg-c3ow">6,491</td>
<td class="tg-c3ow">17,778</td>
</tr>
<tr>
<td class="tg-c3ow">OragnMNIST_Coronal</td>
<td class="tg-c3ow">Abdominal CT</td>
<td class="tg-c3ow">Multi-Class (11)</td>
<td class="tg-c3ow">13,000</td>
<td class="tg-c3ow">2,392</td>
<td class="tg-c3ow">8,268</td>
</tr>
<tr>
<td class="tg-c3ow">OrganMNIST_Sagittal</td>
<td class="tg-c3ow">Abdominal CT</td>
<td class="tg-c3ow">Multi-Class (11)</td>
<td class="tg-c3ow">13,940</td>
<td class="tg-c3ow">2,452</td>
<td class="tg-c3ow">8,829</td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="results">
<h2>Performance Analysis</h2>
<br>
<p>
<center><img src="assets/v1/imgs/performancev2.jpg" border="0" width="100%"></center>
</p>
</div>
<div class="section" id="citation">
<h2>Citation and Licenses</h2>
<p style="text-align:justify; text-justify:inter-ideograph">
If you find this project useful, please cite our ISBI'21 paper as:
<br>
<i>
Jiancheng Yang, Rui Shi, Bingbing Ni. "MedMNIST Classification Decathlon: A Lightweight AutoML
Benchmark for Medical Image Analysis," arXiv preprint arXiv:2010.14925, 2020.
</i>
<br><br>
or using bibtex:
<br>
<i>
@article{medmnist,<br>
title={MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image
Analysis},<br>
author={Yang, Jiancheng and Shi, Rui and Ni,
Bingbing},<br>
journal={arXiv preprint arXiv:2010.14925},<br>
year={2020}<br>
}
</i>
<br>
</p>
<p>
Besides, please cite the corresponding paper if you use any subset of MedMNIST.
Each subset uses the <b>same license</b> as that of the source dataset.
</p>
<h4>PathMNIST</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
Jakob Nikolas Kather, Johannes Krisam, et al., "Predicting survival from colorectal cancer histology
slides using deep learning: A retrospective multicenter study," PLOS Medicine, vol. 16, no. 1, pp.
1–22, 01 2019.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">CC BY 4.0</a>
</i>
<h4>ChestMNIST</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
Xiaosong Wang, Yifan Peng, et al., "Chestx-ray8: Hospital-scale chest x-ray database and benchmarks
on weakly-supervised classification and localization of common thorax diseases," in CVPR, 2017, pp.
3462–3471.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/publicdomain/zero/1.0/" target="_blank">CC0 1.0</a>
</i>
<h4>DermaMNIST</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
Philipp Tschandl, Cliff Rosendahl, and Harald Kittler, "The ham10000 dataset, a large collection of
multisource dermatoscopic images of common pigmented skin lesions," Scientific data, vol. 5, pp.
180161, 2018.
</p>
<p style="text-align:justify; text-justify:inter-ideograph">
Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, and Allan Halpern: “Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)”, 2018; arXiv:1902.03368.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/licenses/by-nc/4.0/" target="_blank">CC BY-NC 4.0</a>
</i>
<h4>OCTMNIST/PneumoniaMNIST</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
Daniel S. Kermany, Michael Goldbaum, et al., "Identifying medical diagnoses and treatable diseases
by image-based deep learning," Cell, vol. 172, no. 5, pp. 1122 – 1131.e9, 2018.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">CC BY 4.0</a>
</i>
<h4>RetinaMNIST</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
DeepDR Diabetic Retinopathy Image Dataset (DeepDRiD), "The 2nd diabetic retinopathy – grading and
image quality estimation challenge," https://isbi.deepdr.org/data.html, 2020.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">CC BY 4.0</a>
</i>
<h4>BreastMNIST</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
Walid Al-Dhabyani, Mohammed Gomaa, Hussien Khaled, and Aly Fahmy, "Dataset of breast ultrasound
images," Data in Brief, vol. 28, pp. 104863, 2020.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">CC BY 4.0</a>
</i>
<h4>OrganMNIST_{Axial,Coronal,Sagittal}</h4>
<p style="text-align:justify; text-justify:inter-ideograph">
Patrick Bilic, Patrick Ferdinand Christ, et al., "The liver tumor segmentation benchmark (lits),"
arXiv preprint arXiv:1901.04056, 2019.
</p>
<p style="text-align:justify; text-justify:inter-ideograph">
Xuanang Xu, Fugen Zhou, et al., "Efficient multiple organ localization in ct image using 3d region
proposal network," IEEE Transactions on Medical Imaging, vol. 38, no. 8, pp. 1885–1898, 2019.
</p>
<i style="text-align:justify; text-justify:inter-ideograph">
<b>License</b>: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">CC BY 4.0</a>
</i>
</div>
</div>
<div class="section">
<p align="center">
Copyright © 2020- MedMNIST Team
</p>
<p align="center">
<small>
Check the source code of this website on <a href="https://github.com/MedMNIST/medmnist.github.io"
target="_blank">GitHub</a>.
</small>
<br>
<small>
This page uses the template of <a href="https://donglaiw.github.io/proj/mitoEM/index.html"
target="_blank">MitoEM</a> from <a href="https://donglaiw.github.io/"
target="_blank">Donglai Wei</a>.
</small>
</p>
</div>
</div>
</body>
</html>