We aim to publish here:
reflections and outcomes from the different workgroups written by volunteers and members of GloBIAS (on a bi-monthly schedule)
reviews on the GloBIAS seminar series, this will be summaries based on transcripts from the YouTube videos (including translations into several languages - we are still expanding the languages - if you want to help us please get in touch via info@globias.org with the subject line blog post translations)
We're thrilled to introduce in more detail a significant initiative to enhance BioImage Analysis training, which started with an in-person workshop and has the aim to update the Data Carpentry BioImage Analysis Curriculum with Python!
Hosted by ISTA in Klosterneuburg, Austria, April 7-8, 2025, the workshop aimed to transform the existing Data Carpentry image processing curriculum into a comprehensive, go-to resource for beginner or intermediate students to learn bioimage analysis with Python for light microscopy data.
The Carpentries project comprises the Software Carpentry, Data Carpentry, and Library Carpentry communities of Instructors, Trainers, Maintainers, helpers, and supporters who share a mission to teach foundational computational and data science skills to researchers. Its target audience is researchers who have little to no prior computational experience, and its lessons are domain specific, building on learners' existing knowledge to enable them to quickly apply skills learned to their own research.
The final updated curriculum will be designed for early-career scientists experienced with biological microscopy images, who possess basic Python knowledge and general biology understanding. This training aims to equip them with the skills to process multiple images in batch, visualize data, perform measurements, and prepare data for statistical analysis, ultimately enabling faster and larger-scale image data processing, reproduction of workflows, opening of proprietary file formats, and improved comprehension of nomenclature in bioimage analysis.
To start this initiative of the workshop brought together 15 selected leading bioimage analysts from across the world, chosen for both their expertise and collaborative mindset. Generously supported by ISTA, the event covered attendance and on-site accommodation. We’ll be sharing the workshop’s outcomes in more detail soon. A first glimpse of achievements so far you can get here.
The two-day workshop featured an intensive program designed for collaborative curriculum development. Day 1 kicked off with Introductions to Carpentries and the existing curriculum by Toby Hodges (Director of Curriculum at The Carpentries) and Kimberly Meechan (maintainer of the Data Carpentry: Image Processing with Python course) followed by project discussions and the formation of working groups to focus on learning objectives, target audience and content discussions. A highlight of the day was an invited lecture by Robert Haase on "Learning and Teaching Bio-image Analysis in the age of AI," leading into dedicated group work on specific projects. Day 2 continued with focused group work and discussions, culminating in presentations of project progress and the future plans for implementation before closing remarks.
This event would not have been possible without the work of Tereza Belinova (ISTA), Marco Dalla Vecchia (ISTA), and Christa Walter (GloBIAS). We look forward to seeing the first outcomes of this course in the coming year.
Get Involved!
We're excited about this initiative and would love for you to be a part of it.
Have questions or want to learn more? Reach out to us at data_carpentry_python@globias.org.
Interested in teaching? We'll be looking for students and teachers to help deliver this curriculum next year. Let us know if you're interested!
Want to contribute right now? You can jump in and help by opening an issue on our repository: https://github.com/carpentries-incubator/bioimage-analysis-python/issues
In the ever-evolving landscape of bioimage analysis, navigating the plethora of available tools and methodologies can be a significant challenge. A recent GloBIAS seminar shed light on the transformative potential of Large Language Models (LLMs) in this domain, featuring an insightful presentation by Caterina Fuster-Barceló from Universidad Carlos III de Madrid on the BioImage.IO chatbot. This innovative tool promises to streamline workflows and enhance accessibility for bioimage analysts worldwide.
Caterina Fuster-Barceló eloquently outlined the exciting possibilities that LLMs, such as ChatGPT, offer for bioimage analysis. These sophisticated models, trained on vast amounts of data, possess the capacity to process diverse information formats and execute complex tasks, potentially revolutionizing how we interact with and analyze biological images.
However, she also highlighted the inherent challenges associated with current LLMs:
Knowledge Cutoff: LLMs may lack awareness of recent advancements published after their last update.
Hallucinations: They can generate factually incorrect information without acknowledging their limitations.
Bias: Training data can introduce biases that affect the model's responses.
Poor Reproducibility: Identical queries might yield different answers, posing challenges for standardized laboratory procedures.
To address these limitations and harness the power of LLMs effectively, the BioImage.IO chatbot has been developed as a collaborative effort and is featured in an article in Nature Methods. Closely linked to the BioImage Model Zoo (bioimage.io), a community-driven repository of deep learning models for bioimage analysis, the chatbot acts as an intelligent assistant designed to enhance the reliability and accessibility of information retrieval.
Key Features of the BioImage.IO Chatbot:
Retrieved Augmented Generation (RAG): This crucial feature underpins the chatbot's reliability by augmenting the LLM's knowledge with a curated community knowledge base comprising documentation from the BioImage Model Zoo and its community partners. User queries are vectorized and compared to vectorized documentation chunks to ensure the retrieved information is relevant and truthful, minimizing hallucinations.
Integration with Online Databases and Services: The chatbot can interact with various online resources, including the BioImage Archive, BioImage Informatics Index, BioTools, Human Protein Atlas, and the Image.sc Forum, via API calls. This allows users to efficiently search for data and information without navigating multiple websites.
Execution of AI Models: The BioImage.IO chatbot has the capability to run specific AI models, such as Cellpose, directly through its interface. This simplifies the process of testing and applying these models to user-provided data.
Extensible Architecture: Built on an extension-based mechanism, the chatbot's functionality can be easily expanded through the development of new extensions in Python or JavaScript. This allows for the integration of diverse tools and workflows, including even real-time microscope control.
User-Centric Design: The chatbot allows users to define their user profile, enabling more tailored and relevant responses based on their expertise (e.g., deep learning developer vs. life scientist). Furthermore, user feedback is actively solicited to continuously improve the chatbot's performance.
To cater to different user needs, the BioImage.IO chatbot offers three distinct assistants:
BioImage Seeker (Melman): Your go-to assistant for searching documentation from community partners and retrieving information via API calls to databases like the BioImage Archive and BioImage Informatics Index.
BioImage Tutor (Bridget): Designed to answer technical questions about bioimage analysis and artificial intelligence, drawing upon a carefully curated library of books and chapters.
BioImage Analyst (Nena): Empowering users to generate and execute code, enabling tasks such as running Cellpose on uploaded images.
The BioImage.IO chatbot is readily accessible through three convenient channels:
As an integrated widget on the BioImage Model Zoo website.
Via a web browser.
As a GPT within the ChatGPT store.
The BioImage.IO chatbot is a testament to the power of community collaboration. Researchers are encouraged to contribute to its ongoing development by creating new extensions, adding documentation for their tools, and providing valuable feedback. This collective effort will ensure that the chatbot remains a relevant and powerful resource for the bioimage analysis community.
The BioImage.IO chatbot represents a significant advancement in leveraging the capabilities of large language models for bioimage analysis. By prioritizing reliable information retrieval, offering extensible functionalities, and maintaining a user-centric approach, it promises to be an invaluable tool for researchers worldwide.
Ready to experience the BioImage.IO chatbot? Explore it through the BioImage Model Zoo website, your web browser, or the ChatGPT store. We encourage you to share your feedback and join the discussion on image.sc to contribute to the continued evolution of this exciting project.
At GloBIAS, we understand that image analysis is a complex task, and that many researchers in the biological field may not be aware of the BioImage Analysts’ community or may not actively engage with it, as often BioImage Analysis just one part of their project, rather than their primary focus. Moreover, we recognize that most searchable image analysis workflows are tailored to specific questions, images, or research contexts. For this reason, we believe it is highly valuable for researchers to have access to the GloBIAS network of professionals, enabling researcher to find the best approach for their unique analyses.
To provide guidance and support, we launched the GloBIAS Call4Help session series, a space where bioimage analysis experts help resolve doubts and answer questions about the use of specific software for image analysis in a specific project context The pilot event took place from January 22nd to February 5th, 2025, meant to allow us to test the set-up for future sessions.
Registration for the next session is open here!
Pilot event
In the pilot event, three software tools were covered: Napari, CellProfiler, and Ilastik. For each software, a total of six sessions were provided, divided into two time slots to accommodate different time zones across the world.
Each session featured two trainers and one attendee, who had the opportunity to receive 40-45 minutes of personalized guidance around the specific question submitted. A GloBIAS moderator was also present to ensure the sessions ran smoothly.
The trainers not only answered questions but also provided recommendations on resources and tutorials for continued learning to enable the attendee to continue working on the bioimage analysis for their project after the session.
Fostering Community Engagement: Leveraging the Image.sc Forum in GloBIAS Call4Help
The pilot session highlighted the importance of submitting questions on the Image.sc forum prior to the event, as it enables better preparation and optimizes the use of available time on both sides. The questions and their answers, posted on the forum, are essential to ensuring they remain accessible for future reference, benefiting a broader community and improving overall information accessibility. To support this, a guide was provided to participants before the sessions, helping them learn how to effectively post their questions on the Image.sc forum.
We also created a guide for the trainers with key recommendations to help them navigate the process smoothly, based on the community experience with online training and advice. Protecting data privacy was a top priority, so participants were encouraged to share representative images instead of full datasets, ensuring confidentiality. Time management was also crucial, a clear time limit was set for each participant, to ensure focused and efficient discussions. However, recognizing that not every question would be fully addressed within the limited time, trainers were prepared with links to relevant tutorials and documentation for further exploration. Additionally, trainers encouraged participants to continue the discussion with the community on their posts in the Image.sc forum, fostering ongoing collaboration and further development of skills.
Impact and Future of Call4Help
The GloBIAS Call4Help pilot event run successfully, with high participation from researchers worldwide. The sessions demonstrated a real need for open access guidance in image analysis, and we, the organisers and as well as the trainers are pleased to have contributed to solving project specific image analysis problems.
Looking ahead, we plan to expand this initiative to include other software and analysis techniques, as well as improve participant selection to ensure an equitable and beneficial experience for everyone.
We thank all the trainers and participants who made this first edition possible. We look forward to seeing you in future sessions!
Registration for the next session is open here!
The field of bioimage analysis is rapidly evolving, and large language models (LLMs) are emerging as powerful tools for researchers and analysts. This blog post summarizes key insights from a recent GloBIAS seminar on LLMs by Robert Haase, offering an introduction tailored for life scientists. The presentation addressed the motivation, applications, and challenges of using LLMs in bioimage analysis, highlighting their potential to revolutionize how we extract meaningful data from biological images.
LLMs are a type of neural network that can be used for various tasks, including translation and code generation. While not originally designed for knowledge or information retrieval, LLMs are increasingly used to answer questions and provide insights. However, it’s important to be aware of potential pitfalls, such as "hallucinations," where the model generates incorrect or misleading information.
Code Generation: LLMs can translate English text into executable code, such as Python, making it easier to automate image analysis tasks.
Image Modification: LLMs can modify microscopy images based on textual instructions, such as blurring or enhancing image quality.
Image Description: LLMs can describe the contents of an image, which is useful for automated image annotation.
One common critique of LLMs is their limited reproducibility, as they may produce different outputs for the same input. However, using LLMs for code generation can mitigate this issue. The generated code, once executed, will consistently produce the same result.
Traditional image processing architectures typically involve encoder-decoder networks, which transform an input image into an output image. LLMs, based on transformer architecture, differ by incorporating three elements: input, output, and a shifted output. This allows LLMs to translate between different forms of data, such as images and text.
Prompt Engineering: Crafting specific prompts to guide the LLM towards the desired output.
Retrieval Augmented Generation (RAG): Enhancing the accuracy of LLMs by integrating them with a domain-specific knowledge base.
Function Calling: Using LLMs to identify and parameterize functions that can perform specific tasks.
Fine-Tuning: Customizing a pre-trained LLM with domain-specific data to improve its performance.
Understand the Code: Always understand the code generated by LLMs and verify its correctness.
Question Established Methods: Use manual measurements to check results.
Disclose LLM Usage: Be transparent about using LLMs in your research methods.
Share Prompts: Share effective prompts with the community to facilitate collective learning.
Large language models hold immense promise for bioimage analysis, offering tools for automation, annotation, and data extraction. By understanding their capabilities and limitations, and by adopting best practices for their use, bioimage analysts can harness the power of LLMs to drive new discoveries and insights.
Explore LLMs for automating routine image analysis tasks.
Experiment with prompt engineering to optimize LLM performance.
Contribute to open-source benchmarks to evaluate and improve LLMs.
Engage with the GloBIAS community to share knowledge and experiences.
This blog post aims to equip you with the knowledge to start exploring and integrating LLMs into your bioimage analysis workflows. The potential of these models is vast, and their responsible application promises to unlock new frontiers in life science research.
(Note: The blog post has been written to be formal yet accessible, suitable for professionals in the field. It incorporates key points, examples, and recommendations from the presentation, and encourages further exploration and community engagement.)
As an imaging scientist in Taiwan for over 14 years, I had been working in isolation without the support of communities, relying primarily on self-learning to develop my skills. However, this began to change in 2024. Bioimage analysts from Taiwan and Japan came together to establish EABIAS, thanks to the efforts of Cheng-Yu, an image analyst studying in the UK. Born in Japan and raised in Taiwan, Cheng-Yu played a key role in connecting us.
With an introduction from “Mr. Japan” Kota Miura and financial support from GloBIAS, I had the honour of attending the first GloBIAS in-person workshop held in Gothenburg.
Before the event, I was assigned to two working groups—one focused on studying the image analysis workflow from an assigned paper, and the other on hosting the career path session. To prepare, I attended two online meetings per week, making this the most intensive pre-workshop preparation I had ever experienced. Yet, it was absolutely worth it! Through these regular online meetings, I connected with bioimage analysts from various countries and also experienced how networking can be fostered through dedicated activities.
For the first time, I felt part of a like-minded community. I also realized that collaboration makes everything more enjoyable. Having spent years promoting image analysis at National Taiwan University, I began to ask myself: How could I reach more people through networking? One thing became clear—we need more bioimage analysts working together in Taiwan.
On my long journey home, I started reaching out to friends working in core facilities across Taiwan, sharing the education materials generously provided by the community and discussing the possibility of organizing a local course series led by us. In our first year, we will start with ImageJ, the most widely used image analysis tool, laying the foundation for transitioning to Python in the second year.
The first local workshop organized by EABIAS—ImageJ Micro-Image Analysis and Programming—will take place every Monday afternoon for seven weeks, starting on March 3rd. The course will be conducted in Chinese, and we still have available slots for online participants. If you prefer learning in your native language, we warmly invite you to join us!
More information:
ImageJ Micro-Image Analysis and Programming (registration now closed)
Over the last few months, the Education and Training group has launched an exciting new initiative: the GloBIAS Seminar Series. This series is designed to support bioimage analysts and the broader community with informative talks that highlight the latest tools, software, and innovations in bioimage analysis. It's also a space for learning, sharing, and connecting. Inspired by initiatives like the Euro-BioImaging Virtual Pub, we recognized a strong interest in having a dedicated platform for presenting and discussing topics in bioimage analysis.
The GloBIAS Seminar Series will feature one seminar during the last week of each month, with presentations scheduled across different time zones to accommodate diverse audiences. Each quarter will focus on a specific theme, allowing us to delve deeper into key topics over three consecutive months. This approach ensures both continuity and a range of perspectives on each topic before moving on to the next.
In the first quarter, we focused on “LLMs for Bioimage Analysis,” featuring presentations by Robert Haase, Caterina Fuster-Barceló, and Loïc Alain Royer. They each showcased unique tools and approaches for integrating large language models (LLMs) into bioimage analysis. Recordings of these sessions are available on our YouTube channel.
Our current theme is Infrastructure for Deploying Image Analysis Workflows. So far, we've had Ankur Kumar present on “Designing and Implementing Systems to Process Petabyte-Scale Imaging Datasets,” and Christian Tischer discussing “NextFlow for Bioimage Analysis.” Coming up, we’re excited to have Beatriz Serrano-Solano join us on Wednesday, February 26, 2025, from 2:00 pm to 3:30 pm CET to discuss “Image Analysis Using Galaxy.” This session promises to be an excellent addition to our line-up!
We encourage presenters and participants alike to continue any unanswered questions or discussions in the Image.sc forum, a platform open to the entire community. Ongoing conversations led by Robert Haase and Caterina Fuster-Barceló can be found here and here.
There are many more topics on the horizon, and we invite you to join the workgroup, suggest ideas, or even propose talks. Thank you all of you for being part of the journey!
Don’t forget to register!