Research Assistant @HochschuleFL | MSc @bhupro | AI🤖 in pathology🔬🦠🧬🧫 | Coffee☕ | Nature🏞 | Dogs🐕‍🦺 | Let's geek out!

Flensburg, Schleswig-Holstein
Joined July 2020
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
🚨 New publication alert: 📢 “Benchmarking Deep Learning and Vision Foundation Models for Atypical vs. Normal Mitosis Classification with Cross-Dataset Evaluation.” 🖊️ @SwetaBioX, V Weiss, T A Donovan, R HJ Fick, T Conrad, ..., K Breininger, M Aubreville, C A Bertram. ⬇️
1
1
1
199
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
My manuscripts folder
28
62
11
1,173
108,303
This is a very useful resource!
Today is the day! 🎉 After 1.5 years of hard work, "Build A Large Language Model (From Scratch)" is finally out! Print or ebook copies are available on Manning’s site: mng.bz/M96o 📚. And it's also available on Amazon soon: amazon.com/Build-Large-Langu… 📦.
126
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Iranian chess player Dorsa Derakhshani was given an ultimatum by the Islamic regime: wear the hijab or you won’t be allowed to compete. Not only she refused to wear the hijab, she joined the USA team and became a world champion. Queen behavior if you ask me. 👑
1,877
22,546
1,014
222,548
3,983,086
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
If pasta, bread, and red meat are so bad, why do Europeans live to 100+ and stay healthy? But when Americans consume these? They get chronically sick. After some research… I found the disturbing truth that explains exactly what's going on: 🧵
1,096
8,555
841
40,440
16,362,143
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Thread of cool maps you've (probably) never seen before 🧵 1. All roads lead to Rome
1,521
37,550
3,282
494,964
63,180,336
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
I am seeing a lot of companies that want to solve biology with AI and foundation models. Some examples include: Isomorphic Labs, EvolutionaryScale, DiffuseBio, Xaira Therapeutics, Recursion, Inceptive, etc. But I feel like I don't see as many companies trying to solve healthcare with AI. Sure there are medical AI companies like Abridge, Nabla, Hippocratic AI, etc. and they're all doing great work. However they are often limited to narrow administrative use-cases of LLMs, and do not seem to aim for more ambitious and innovative use-cases of foundation models in healthcare. Sure, administrative use-cases like clinical transcription/notetaking are very useful and important low-hanging fruits for AI applications, but there's so much more that foundation models can do to revolutionize healthcare. And I do not see many companies doing so. Of course the reason why is completely obvious: healthcare is a difficult industry to disrupt due to regulations and scale and data and so many more challenges. But just because it's hard doesn't mean it's not worth trying, right? Would love to hear others thoughts on this and if other people are working on this...
60
39
3
386
103,695
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Science is not really happening when the goal is a paper, a presentation, a grant, or any product. It only really happens when the goal is a process of discovery.
51
625
72
3,579
231,160
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
ACCV'24: 7 days. WACV'25-R1: 17 days. 3DV'25: 45 days. AAAI'25: 47 days. WACV'25-R2: 73 days. ICASSP'25: 73 days. ICRA'25: 79 days.
2
18
2,301
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Thanks for your fantastic presentations to @florianjug, @cppape & all students! 👌We had a blast today @BIGS @uniallianz at #ISAS & enjoyed the fruitful discussions about #bioimage analyses with all of you. 🔬 🤖 Talk soon 👋 @JianxuChen @JustinSonneck
3
13
1,022
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Found my father's BSc Mathematics books which are more than 50 years old (1972-75) Completely yellowed but still lush with notes and in an amazing condition - no eBooks and iPad notes can beat this feeling of preservation.
58
147
19
2,549
89,463
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
What is your favorite textbook (in any field)? Why is it your favorite?
60
29
12
170
94,898
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Have you registered for #BIGS @uniallianz on Friday (June 21), yet? It's about #biomedical image analyses. If you are a young researcher & like to learn from i.e. @florianjug & @cppape: isas.de/en/events/bigs-biome… The event is free of charge, in Dortmund & online.@JianxuChen @axlmsg
4
1
9
1,894
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
🚨 #ChatGPT + @napari_imaging 🚨 Releasing my latest weekend project: Omega, an autonomous LLM agent that writes image processing and analysis code, fixes its mistakes, accesses the napari viewer, makes widgets, & more! github.com/royerlab/napari-c… @langchain @OpenAI #OmegaAgent
14
225
51
814
328,968
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
What massively improved your mental health?
6,271
1,339
7,467
21,164
11,306,900
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
Copy from one paper = Plagiarism Copy from several papers = Literature review 🤣🤣🤣
105
7,569
380
31,843
1,355,199
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
To give context on age, Grace Hopper began computing at 38, completed the first compiler at 46, helped shape COBOL at 53, kept developing COBOL for the Navy in her 70s, retired from the Navy at 80 & then became a consultant for the Digital Equipment Corporation. v/@cooperx86
20
335
27
1,193
142,727
Sweta Banerjee (শ্বেতা ব্যানার্জী) retweeted
One of my graduate students dropped by my office, exasperated, and said, "I don't know how to innovate. So, how can I do a PhD?". I erased my whiteboard and began listing strategies for generating new research ideas. They left reassured and excited! As #professors, we mentor and make knowledge accessible to empower our students. I hope this is helpful to others! #academicchatter #Phdlife #Highered
26
624
29
2,879
335,565
This is an interesting thread 🤔
C elegans has 302 neurons. Show me an artificial neural net with 302 units that can do everything a worm can.
117