-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
517 lines (473 loc) · 22.9 KB
/
Copy pathindex.html
File metadata and controls
517 lines (473 loc) · 22.9 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
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Embodied Clinical Intelligence Colloquium (ECLIC)</title>
<meta name="description" content="Embodied Clinical Intelligence establishes a new paradigm where AI systems are situated in clinical environments, interact with clinicians, and assist in real-time clinical workflows. A research community advancing AI agents in medicine.">
<meta name="keywords" content="embodied clinical intelligence, AI in medicine, clinical AI agents, medical AI research, AI clinical workflows, clinical decision support, healthcare AI, AI in radiology, large language models medicine, reinforcement learning clinical">
<meta name="color-scheme" content="light dark">
<meta name="robots" content="index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1">
<link rel="icon" href="favicon.ico" sizes="any" />
<link rel="icon" type="image/png" href="favicon.png?v=3" />
<link rel="apple-touch-icon" href="apple-touch-icon-v2.png" />
<link rel="canonical" href="https://embodied-clinical-intelligence.github.io">
<link rel="sitemap" type="application/xml" href="/sitemap.xml">
<!-- Open Graph -->
<meta property="og:title" content="Embodied Clinical Intelligence Colloquium (ECLIC)">
<meta property="og:description" content="A research community advancing embodied AI systems in medicine. Unapolegetically technical, uncompromisingly clinical.">
<meta property="og:type" content="website">
<meta property="og:url" content="https://embodied-clinical-intelligence.github.io">
<meta property="og:image" content="https://embodied-clinical-intelligence.github.io/favicon.png">
<meta property="og:site_name" content="Embodied Clinical Intelligence Colloquium">
<meta property="og:locale" content="en_US">
<!-- Twitter Card -->
<meta name="twitter:card" content="summary">
<meta name="twitter:title" content="Embodied Clinical Intelligence Colloquium (ECLIC)">
<meta name="twitter:description" content="Research community advancing embodied AI systems in clinical medicine — agents, workflows, and decision support.">
<meta name="twitter:image" content="https://embodied-clinical-intelligence.github.io/favicon.png">
<!-- Structured Data -->
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "ResearchOrganization",
"name": "Embodied Clinical Intelligence Colloquium",
"alternateName": "ECLIC",
"url": "https://embodied-clinical-intelligence.github.io",
"logo": "https://embodied-clinical-intelligence.github.io/favicon.png",
"description": "A research community advancing embodied clinical intelligence — AI systems situated in clinical environments that interact with clinicians and assist in real-time clinical workflows and decision-making.",
"knowsAbout": [
"Embodied Clinical Intelligence",
"Artificial Intelligence in Medicine",
"Clinical AI Agents",
"Reinforcement Learning",
"Vision-Language Models",
"Medical AI Evaluation",
"Clinical Decision Support Systems",
"AI in Radiology"
],
"event": {
"@type": "EducationEvent",
"name": "ECLIC Workshop Series",
"description": "Workshop series covering technical AI and clinical topics at the intersection of embodied intelligence and medicine",
"organizer": {
"@type": "Organization",
"name": "Embodied Clinical Intelligence Colloquium"
},
"eventAttendanceMode": "https://schema.org/MixedEventAttendanceMode",
"about": [
"Reinforcement Learning for Clinical AI",
"Vision-Language Models in Medicine",
"Clinical AI Evaluation",
"AI Agents in Healthcare"
]
}
}
</script>
<link rel="preconnect" href="https://fonts.googleapis.com" />
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
<link href="https://fonts.googleapis.com/css2?family=Spline+Sans+Mono:wght@400;500;550;600;700&family=Manrope:wght@400;500&display=swap" rel="stylesheet" />
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.5.2/css/all.min.css" integrity="sha512-SnH5WK+bZxgPHs44uWIX+LLJAJ9/2PkPKZ5QiAj6Ta86w+fsb2TkcmfRyVX3pBnMFcV7oQPJkl9QevSCWr3W6A==" crossorigin="anonymous" referrerpolicy="no-referrer" />
<style>
:root {
--bg: #ffffff;
--fg: #000000;
--muted: #57606a;
--accent: #1f6feb;
--card-border: rgba(13, 17, 23, 0.08);
--maxw: 960px;
--padx: 40px;
}
@media (prefers-color-scheme: dark) {
:root {
--bg: #0d1117;
--fg: #c9d1d9;
--muted: #8b949e;
--accent: #58a6ff;
--card-border: rgba(255, 255, 255, 0.25);
}
}
html[data-theme="dark"] {
--bg: #0d1117;
--fg: #c9d1d9;
--muted: #8b949e;
--accent: #58a6ff;
--card-border: rgba(255, 255, 255, 0.25);
}
html[data-theme="light"] {
--bg: #ffffff;
--fg: #000000;
--muted: #57606a;
--accent: #1f6feb;
--card-border: rgba(13, 17, 23, 0.08);
}
* { margin: 0; padding: 0; box-sizing: border-box; }
html {
overflow-y: scroll;
scrollbar-gutter: stable;
}
html, body { height: 100%; }
body {
font: 16px/1.7 "Manrope", -apple-system, BlinkMacSystemFont, "Segoe UI", Helvetica, Arial, sans-serif;
color: var(--fg);
background: var(--bg);
-webkit-font-smoothing: antialiased;
-moz-osx-font-smoothing: grayscale;
}
a {
color: var(--fg);
text-decoration: underline;
}
a:hover { text-decoration: none; }
.theme-toggle {
position: fixed;
top: 20px;
right: 20px;
z-index: 100;
background: transparent;
border: 1px solid var(--card-border);
border-radius: 50%;
width: 36px;
height: 36px;
display: flex;
align-items: center;
justify-content: center;
cursor: pointer;
color: var(--fg);
font-size: 18px;
transition: border-color 0.2s ease;
}
.theme-toggle:hover {
border-color: var(--fg);
}
.ascii-video-wrap {
position: fixed;
top: 0;
right: 0;
bottom: 0;
left: calc(300px + var(--maxw) - 300px);
z-index: 0;
pointer-events: none;
user-select: none;
display: flex;
align-items: center;
justify-content: center;
overflow: hidden;
/* Opaque bg matching the page so the video's flat background can
blend into it (mix-blend-mode blends within this stacking
context, not the whole page). Makes the video rectangle seamless. */
background: var(--bg);
}
.ascii-video-wrap video {
max-width: 780px;
max-height: 850px;
object-fit: contain;
opacity: 1;
/* Blend the video's flat background into the page so its rectangle
edge is mathematically invisible regardless of color management:
white bg + multiply (light), black bg + screen (dark). */
mix-blend-mode: multiply;
}
@media (prefers-color-scheme: dark) {
.ascii-video-wrap video { mix-blend-mode: screen; }
}
html[data-theme="dark"] .ascii-video-wrap video { mix-blend-mode: screen; }
html[data-theme="light"] .ascii-video-wrap video { mix-blend-mode: multiply; }
@media (max-width: 1400px) {
/* Stack the layout: video on top, content below. The <aside> follows
<main> in the DOM, so flex ordering lifts it above the title.
Override the base `height:100%` to auto, else the fixed-height
flex column squishes the video into a sliver. */
body {
display: flex;
flex-direction: column;
height: auto;
min-height: 100%;
}
.ascii-video-wrap {
order: -1;
flex-shrink: 0;
position: relative;
left: 0;
top: auto;
right: auto;
bottom: auto;
width: 100%;
justify-content: center;
padding: 24px 0 8px;
}
.ascii-video-wrap video {
max-width: 450px;
max-height: 560px;
}
}
.wrap {
max-width: var(--maxw);
margin: 0;
padding: 60px 40px 80px 240px;
position: relative;
z-index: 2;
}
@media (max-width: 1400px) {
.wrap {
padding: 40px 48px 48px;
margin: 0 auto;
}
.wrap h1 { margin-top: 8px; margin-bottom: 48px; text-align: center; }
.wrap footer { text-align: center; margin-top: 12px; padding-bottom: 24px; }
}
h1 {
font-family: "Spline Sans Mono", monospace;
font-weight: 500;
font-size: clamp(32px, 5vw, 42px);
line-height: 1.2;
margin-bottom: 40px;
letter-spacing: -0.3px;
}
hr {
border: none;
height: 1px;
background: var(--card-border);
margin: 48px 0;
}
h2 {
font-family: "Spline Sans Mono", monospace;
font-weight: 500;
font-size: clamp(22px, 3vw, 28px);
letter-spacing: -0.3px;
color: var(--fg);
margin-bottom: 20px;
}
.vision {
font-size: 1.05rem;
line-height: 1.8;
color: var(--fg);
text-align: left;
}
.vision + .vision {
margin-top: 24px;
}
.schedule {
margin-top: 0;
}
.session {
padding: 24px 0;
}
.session + .session {
border-top: 1px solid var(--card-border);
}
.session-label {
font-size: 12px;
font-weight: 500;
text-transform: uppercase;
letter-spacing: 0.5px;
margin-bottom: 4px;
}
.session-label.technical { color: #ef4444; }
.session-label.clinical { color: #3b82f6; }
/* Grey out in dark mode (OS preference), but only when the user hasn't
explicitly forced a theme via the toggle. */
@media (prefers-color-scheme: dark) {
html:not([data-theme]) .session-label.technical,
html:not([data-theme]) .session-label.clinical { color: var(--muted); }
}
html[data-theme="dark"] .session-label.technical,
html[data-theme="dark"] .session-label.clinical { color: var(--muted); }
/* Explicit light toggle always restores the accent colors, even when
the OS is in dark mode. */
html[data-theme="light"] .session-label.technical { color: #ef4444; }
html[data-theme="light"] .session-label.clinical { color: #3b82f6; }
.session-title {
font-family: "Manrope", -apple-system, BlinkMacSystemFont, sans-serif;
font-weight: 500;
font-size: 1.05rem;
color: var(--fg);
margin-bottom: 10px;
}
.session-readings {
list-style: none;
padding: 0;
margin: 0;
}
.session-readings li {
font-size: 14px;
color: var(--muted);
line-height: 1.6;
padding: 2px 0 2px 16px;
position: relative;
}
.session-readings li::before {
content: '';
position: absolute;
left: 0;
top: 10px;
width: 4px;
height: 4px;
border-radius: 50%;
background: var(--card-border);
}
.session-readings li a {
color: var(--muted);
}
footer {
margin-top: 16px;
color: var(--muted);
font-size: 13px;
text-align: left;
padding: 16px 0 48px;
}
@media (max-width: 600px) {
.wrap { padding: 40px 32px 12px; }
}
</style>
</head>
<body>
<button class="theme-toggle" id="theme-toggle" aria-label="Toggle dark mode">
<i class="fa-solid fa-moon" id="theme-icon"></i>
</button>
<main class="wrap">
<header>
<h1>Embodied Clinical Intelligence Colloquium (ECLIC)</h1>
</header>
<section aria-label="About">
<p class="vision">
AI systems are currently siloed into highly-specific clinical tasks and are not suitable to handle the complexities and nuances of clinical care.
<strong>Embodied Clinical Intelligence</strong> establishes a new paradigm in which AI systems ground themselves in clinical context, act within real-world clinical workflows,
and interact with clinicians in real-time.
</p>
<p class="vision">
We are a community of researchers at the intersection of AI and medicine that are <strong>unapologetically technical and uncompromisingly clinical</strong>. In our mission of pushing the frontiers of research in embodied clinical intelligence,
we host workshops, fireside chats, and guest lectures, to build the community advancing the next generation of embodied AI systems in medicine.
</p>
</section>
<hr />
<section aria-label="Workshop Series">
<h2>Workshop Series</h2>
<div class="schedule">
<div class="session">
<div class="session-label technical">Technical</div>
<div class="session-title">Session 1: Beyond the OpenClaw Hype: From Reinforcement Learning to ReAct</div>
<ul class="session-readings">
<li><a href="http://incompleteideas.net/book/the-book.html">Reinforcement Learning: An Introduction (Sutton and Barto) [1992]</a></li>
<li><a href="https://arxiv.org/abs/2210.03629">ReAct: Synergizing Reasoning and Acting in Language Models [2022]</a></li>
<li><a href="https://arxiv.org/abs/2303.11366">Reflexion: Language Agents with Verbal Reinforcement Learning [2023]</a></li>
<li><a href="https://arxiv.org/abs/2405.15793">SWE-Agent: Agent-Computer Interfaces Enable Automated Software Engineering [2024]</a></li>
</ul>
</div>
<div class="session">
<div class="session-label clinical">Clinical</div>
<div class="session-title">Session 2: Why We Can't Just Throw AI Algorithms at the EHR: The Messy Reality of Clinical Data</div>
<ul class="session-readings">
<li><a href="https://doi.org/10.1148/ryai.210315">RadImageNet: An Open Radiologic Deep Learning Research Dataset for Effective Transfer Learning [2022]</a></li>
<li><a href="https://doi.org/10.1038/s41591-022-01981-2">Multimodal Biomedical AI [2022]</a></li>
</ul>
</div>
<div class="session">
<div class="session-label technical">Technical</div>
<div class="session-title">Session 3: Vision-Language Models</div>
<ul class="session-readings">
<li><a href="https://arxiv.org/abs/2103.00020">Learning Transferable Visual Models From Natural Language Supervision (CLIP) [2021]</a></li>
<li><a href="https://arxiv.org/abs/2304.08485">Visual Instruction Tuning (LLaVA) [2023]</a></li>
</ul>
</div>
<div class="session">
<div class="session-label clinical">Clinical</div>
<div class="session-title">Session 4: A Brief History of Clinical AI Evaluation: The Pitfalls of Automated Evaluation Metrics</div>
<ul class="session-readings">
<li><a href="https://arxiv.org/abs/1705.02315">ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases [2017]</a></li>
<li><a href="https://doi.org/10.1016/j.patter.2023.100802">Evaluating progress in automatic chest X-ray radiology report generation (RadCliQ) [2023]</a></li>
</ul>
</div>
<div class="session">
<div class="session-label technical">Technical</div>
<div class="session-title">Session 5: Reinforcement Learning with Large Language Models</div>
<ul class="session-readings">
<li><a href="https://link.springer.com/article/10.1007/BF00992696">REINFORCE: Simple statistical gradient-following algorithms for connectionist reinforcement learning [1992]</a></li>
<li><a href="https://arxiv.org/abs/2203.02155">PPO: Training Language Models to Follow Instructions with Human Feedback (RLHF) [2022]</a></li>
<li><a href="https://arxiv.org/abs/2402.03300">GRPO: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models [2024]</a></li>
<li><a href="https://arxiv.org/abs/2501.12948">DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning [2025]</a></li>
</ul>
</div>
<div class="session">
<div class="session-label clinical">Clinical</div>
<div class="session-title">Session 6: Lessons Learned from Running Clinical Reader Studies</div>
<ul class="session-readings">
<li><a href="https://doi.org/10.1186/s12880-024-01435-w">An Open-source Fine-tuned Large Language Model for Radiological Impression Generation: A Multi-Reader Performance Study [2024]</a></li>
<li>Radiologically Relevant Clinical History Summarization With Large Language Models: A Multi-Reader Performance Study [2026]</li>
</ul>
</div>
<div class="session">
<div class="session-label technical">Technical</div>
<div class="session-title">Session 7: Developing Agents from a Reinforcement Learning Perspective</div>
<ul class="session-readings">
<li><a href="https://doi.org/10.1016/S0004-3702(98)00023-X">Planning and Acting In Partially Observable Stochastic Domains (POMDP) [1998]</a></li>
<li>RL Environments: <a href="https://huggingface.co/blog/openenv">OpenEnv</a>, <a href="https://www.primeintellect.ai/">Prime Intellect</a></li>
<li><a href="https://arxiv.org/abs/2404.07972">OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments [2024]</a></li>
<li><a href="https://arxiv.org/abs/2501.12326">UI-TARS: Pioneering Automated GUI Interaction with Native Agents [2025]</a></li>
<li><a href="https://arxiv.org/abs/2508.09123">OPENCUA: Open Foundations for Computer-Use Agents [2025]</a></li>
</ul>
</div>
<div class="session">
<div class="session-label clinical">Clinical</div>
<div class="session-title">Session 8: Why Computer Scientists Need to Shadow Clinicians</div>
<ul class="session-readings">
<li>Case Study: Hallucinated Comparisons in Radiology Report Generation Vision-Language Models</li>
<li>Experience shadowing Dr. Jae Ho Sohn in the reading room</li>
</ul>
</div>
<div class="session">
<div class="session-title">Conclusion: Towards Embodied Clinical Intelligence</div>
<ul class="session-readings">
<li><a href="https://doi.org/10.1038/s41746-024-01083-y">Evaluating Large Language Models as Agents in the Clinic [2024]</a></li>
<li><a href="https://doi.org/10.1038/s41586-025-08866-7">Towards Conversational Diagnostic Artificial Intelligence (AMIE) [2025]</a></li>
<li>EHRGym: Training and Evaluating Computer-Use Agents in Clinical Workflow Environments Towards Embodied Clinical Intelligence</li>
<li>Future Directions</li>
</ul>
</div>
</div>
</section>
<footer style="user-select:none;">
<span>© <span id="year"></span> Embodied Clinical Intelligence</span>
</footer>
</main>
<aside class="ascii-video-wrap" aria-hidden="true">
<video id="ascii-video" autoplay muted loop playsinline aria-label="ASCII art animation of a clinical robot representing embodied clinical intelligence"></video>
</aside>
<script>
document.getElementById('year').textContent = new Date().getFullYear();
(function() {
var video = document.getElementById('ascii-video');
var toggle = document.getElementById('theme-toggle');
var icon = document.getElementById('theme-icon');
var mq = window.matchMedia('(prefers-color-scheme: dark)');
var currentSrc = '';
function isDark() {
var theme = document.documentElement.getAttribute('data-theme');
if (theme) return theme === 'dark';
return mq.matches;
}
function apply() {
var dark = isDark();
document.body.style.background = dark ? '#0d1117' : '#ffffff';
icon.className = dark ? 'fa-solid fa-sun' : 'fa-solid fa-moon';
var src = dark ? 'ascii_dark.mp4' : 'ascii_light.mp4';
if (src !== currentSrc) {
currentSrc = src;
video.src = src;
video.load();
video.play();
}
}
toggle.addEventListener('click', function() {
var dark = isDark();
document.documentElement.setAttribute('data-theme', dark ? 'light' : 'dark');
apply();
});
mq.addEventListener('change', function() {
if (!document.documentElement.getAttribute('data-theme')) apply();
});
apply();
})();
</script>
</body>
</html>