alings with main branch #7

SciKit-Surgery · Nov 3, 2023 · a985838 · a985838
1 parent 697bb51
commit a985838
Show file tree

Hide file tree

Showing 7 changed files with 148 additions and 1 deletion.
diff --git a/README.md b/README.md
@@ -0,0 +1,60 @@
+# :nut_and_bolt: Hacking Real-time AI workflows for Surgery :wrench: 
+Hackathon hosted by CMIC    
+9-10th November 2023    
+https://cmic-ucl.github.io/CMICHACKS/    
+
+## :scroll: Project description
+Artificial Intelligence (AI)-based surgical workflows use input from multiple sources of data (e.g., medical devices, trackers, robots, cameras, etc) with different Machine Learning and Deep Learning tasks (e.g., classification, segmentation, synthesis, etc) and have been applied across different surgical and acquisition workflows.
+However, the diversity of data sources, pre-processing methods, training and inference methods make a challenging scenario for low-latency applications in surgery.
+In this hackathon, instructors aim to engage with participants to bring three learning outcomes:
+* participants will learn to train, optimise, test and deploy AI models for detection and tool tracking, 
+* participants will learn good software practices to contribute to our open-source projects aligning with medical device software standard (IEC 62304), and 
+* participants will have the chance to work with NVIDIA Clara AGX - A Universal computing architecture for next-generation AI medical instruments.    
+
+We hope to bring together researchers, engineers and clinicians across different departments to hack workflows for real-time AI for Surgery (development, evaluation and integration) and hopefully to sparking future collaborations.
+
+
+## :sparkles: Project leaders and organisers 
+Leaders: [Zhehua Mao](https://github.com/ZH-Mao) and [Miguel Xochicale](https://github.com/mxochicale)    
+Co-organisers: Steve Thompson, Tom Dowrick, Sophia Bano, Matt Clarkson at UCL; and Mikael Brudfors at NVIDIA
+
+## :date: Agenda
+The following is a general agenda that might have slightly changes.
+
+### Thursday 9th Nov 2023 
+| Time  | Activity                                                            |
+| --- |---------------------------------------------------------------------|
+| 9:00 – 10:00 | Setting up                                                          |
+| 10:00 – 11:00 | Welcome session                                                     |
+| 11:00 -12:00 | Hacking: Intro and preparation to hack [:link:](program/day-one/README.md#1100--1200--intro-and-preparations-to-hack)                   |
+| 12:00 - 13:00 | Lunch                                                               | 
+| 13:00 - 15:00 | Hacking: Get familiar with your data [:link:](program/day-one/README.md#1300---1500--get-familiar-with-your-data) |
+| 15:00 - 15:15 | Coffee break                                                        | 
+| 15:15 - 17:00 | Hacking: Training AI workflow [:link:](program/day-one/README.md#1515---1700--hacking-training-ai-workflow)                                      | 
+| 17:00 – all night | Social                                                              |
+
+### Friday 10th Nov 2023 
+| Time  | Activity                                                                        |
+| --- |---------------------------------------------------------------------------------|
+| 09:00 - 10:30 | Hacking: Evaluation of AI models with benchmarks [:link:](program/day-two/README.md) |
+| 10:30 - 10:45 | Coffee break                                                                    | 
+| 10:45 - 12:00 | Hacking: Optmise, test and deploy AI models in clara-agx [:link:](program/day-two/README.md)                       |
+| 12:00 - 13:00 | Lunch                                                                           |
+| 13:00 - 14:30 | Hacking: Deploying models to nvidia-clara-agx [w/Mikael Brudfors] [:link:](program/day-two/README.md)              | 
+| 14:30 – 14:45 | Coffee break                                                                    |
+| 14:45 - 16:30 | Hacking: Tidied up documentation and present results [w/Mikael Brudfors]  [:link:](program/day-two/README.md)      | 
+| 16:30 - 17:00 | Closing session                                                                 |
+
+## :school: Registration and venue
+Registration is free and you can get your tickets [here](https://www.eventbrite.co.uk/e/cmichacks-2-tickets-733294452447)
+The venue of the hackathon is Jeffery (Thursday) and Elvin (Friday) rooms, Level 1, main UCL Institute of Education (IOE) building, 20 Bedford Way, London WC1H 0AL.    
+
+## :computer: Pre-requisites for participants  
+* Participants require to bring their own computers either with CPU or GPU. Computers with GPU will be idea to prototype models. However, we will also prepare pre-training models and simple AI workflows that does not require GPUs. 
+* Have a Github username. Otherwise, [sign up](https://docs.github.com/en/get-started/signing-up-for-github/signing-up-for-a-new-github-account) to GitHub to create an account.
+* Clone https://github.com/SciKit-Surgery/cmicHACKS2.git  
+
+## :family: Forum
+We are using [Discussions](https://github.com/SciKit-Surgery/cmicHACKS2/discussions/) as a place to connect with other members of our community.
+Ask questions you’re wondering about, share ideas, engage with other community members, welcome others and are open-minded. 
+Remember that this is a community we build together 💪!
diff --git a/program/README.md b/program/README.md
@@ -0,0 +1,2 @@
+# Program
+
diff --git a/program/day-one/README.md b/program/day-one/README.md
@@ -0,0 +1,68 @@
+# Thursday 9th Nov 2023
+
+## 11:00 -12:00 > Intro and preparations to hack
+> Hacking: Intro the full AI pipeline and Setting GitHub repository: clone repo, GitHub workflow to code together and document as you hack
+
+1. Setup your [Personal Access Token](https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/managing-your-personal-access-tokens)
+2. Clone repo
+```
+git clone https://github.com/SciKit-Surgery/cmicHACKS2.git 
+```
+3. Workflow for issue management 
+```mermaid
+  flowchart TD;
+      Z[Bug Reported] -->A[...];  
+      A[Bug resolution] -->B(Testing OK?);
+      B--Yes-->C[Prepare commit];
+      B--No-->D[Reopen issue];
+      C ----> E[Push Documentation];
+      D----> A[Bug resolution];
+      E --> F[Close Bug Issue]
+```
+
+## 13:00 - 15:00 > Get familiar with your data
+> Get familiar with your data (download data, label data, etc.)
+
+![fig](kvasir-seg-dataset.png)
+
+1. Download Kvasir-SEG dataset
+> The Kvasir-SEG dataset (size 46.2 MB) contains 1000 polyp images and their corresponding ground truth from the Kvasir Dataset v2. 
+> The resolution of the images contained in Kvasir-SEG varies from 332x487 to 1920x1072 pixels. 
+> The images and its corresponding masks are stored in two separate folders with the same filename. 
+> The image files are encoded using JPEG compression, and online browsing is facilitated. 
+> The bounding box (coordinate points) for the corresponding images are stored in a JSON file.
+https://datasets.simula.no/downloads/kvasir-seg.zip
+
+
+2. Label data. 
+See references for label tools
+
+## 15:15 - 17:00 > Hacking: Training AI workflow
+> Hacking: Training AI workflow   
+![fig](ml-pipeline.png)
+
+* train Segmentation models
+* convert `checkpoint.pth` to `colon.onnx"`
+
+## References 
+
+Jha, Debesh, Pia H. Smedsrud, Michael A. Riegler, Pål Halvorsen, Thomas de Lange, Dag Johansen, and Håvard D. Johansen, 
+"Kvasir-seg: A segmented polyp dataset" 
+Proceedings of the International Conference on Multimedia Modeling, pp. 451-462, 2020.
+[google-citations](https://scholar.google.com/scholar?cites=15924410051330387241&as_sdt=2005&sciodt=0,5&hl=en)
+[dataset](https://datasets.simula.no/kvasir-seg/)
+
+Jha D, Ali S, Tomar NK, Johansen HD, Johansen D, Rittscher J, Riegler MA, Halvorsen P. 
+Real-Time Polyp Detection, Localization and Segmentation in Colonoscopy Using Deep Learning
+IEEE Access. 2021 Mar 4;9:40496-40510. doi: 10.1109/ACCESS.2021.3063716. PMID: 33747684; PMCID: PMC7968127.
+[google-citations](https://scholar.google.com/scholar?cites=11882550127852592683&as_sdt=2005&sciodt=0,5&hl=en)
+
+Label tools:   
+* https://www.robots.ox.ac.uk/~vgg/software/via/  
+* https://labelstud.io/  
+* https://github.com/wkentaro/labelme
+* https://github.com/opencv/cvat
+* https://github.com/microsoft/VoTT
+
+Other open datasets
+* https://datasets.simula.no/ 
diff --git a/program/day-one/kvasir-seg-dataset.png b/program/day-one/kvasir-seg-dataset.png
diff --git a/program/day-one/ml-pipeline.png b/program/day-one/ml-pipeline.png
diff --git a/program/day-two/README.md b/program/day-two/README.md
@@ -0,0 +1,16 @@
+# Friday 10th Nov 2023
+
+## 09:00 - 10:30 > Hacking: Evaluation of AI models with benchmarks
+* ...
+
+## 10:45 - 12:00 > Hacking: Optmise, test and deploy AI models in clara-agx
+* ...
+
+## 13:00 - 14:30 > Hacking: Deploying tested models to nvidia-clara-agx 
+* ...
+
+## 14:45 - 16:30 > Tidied up documentation and present results
+* ... 
+
+
+
diff --git a/slides/README.md b/slides/README.md
@@ -48,7 +48,8 @@ rm -rf .jekyll-cache/ _site/ Gemfile.lock
 1. Setting up pages at https://github.com/SciKit-Surgery/cmicHACKS2/settings/pages
 2. Select deploy from a branch, select branch `7-slides` with path `/root` and[SAVE]. Then, you might need to select `main` branch for the final version of the slides.
 3. First GitHub action:  https://github.com/SciKit-Surgery/cmicHACKS2/actions/runs/6747753816
-4. Online slides should be available here https://scikit-surgery.github.io/cmicHACKS2/slides.html
+4. Online slides should be available here https://scikit-surgery.github.io/cmicHACKS2/slides/slides.html
+
 
 ## References 
 * https://github.com/oss-for-surgtech/ucl-open-science-awards-2023/tree/main/slides