GSoC Week 5

From LH-RIS Viewer to AI Model Service


Last week, I primarily focused on coding and making updates to the LH-RIS (Libre Health Radiology Information System) viewer. This week, my work shifted towards the LH AI Model Service.

The current architecture of the Flask REST-API within the AI Model Service was unfamiliar to me. It was built in an object-oriented style, which seemed complex compared to the simpler implementations I've encountered in the past. As such, I took my time to understand the system's components before implementation.

Implementing the Model Selection API


Once I grasped the system, I dived into implementing the Model Selection API. We introduced a few methods on the viewer side that now transmit Dicom tags information and the new model configuration settings, input by the admin in the administration settings. This information is crucial for the Model Selection API, as it determines which model to invoke on the fly.

The concept behind the Model Selection API is straightforward: It offers a flexible endpoint enabling the switch between different ML models dynamically. This approach eliminates the need to hardcode the model choice in the application, as the Model Selection API uses the model argument in the API request instead.

The Model Selection Algorithm in Action


The model selection algorithm follows these main steps:

- Parse the DICOM tags, model properties, and other information relevant to model selection.
- Analyze the DICOM tags information to see if specific tags suggest that the image belongs to a particular modality. (For example: The attribute "BodyPartExamined" with a value "CHEST" would suggest a chest x-ray.)
- If the tags suggest a certain modality, the algorithm filters the model properties for that specific modality.
- With the "task" and "endpoint" properties of the filtered model properties configurations, the algorithm can call the corresponding model. It creates form data and file data (image) and makes a POST request to it.
- By implementing this way, there was no need to modify existing model methods like cheXnet. After testing the implementation on the AI Model Services and debugging a few errors, the output from the Postman call worked flawlessly.

Integration Testing and Future Plans


For the final part of the process, I ran both Docker images with the updated code to ensure the model selection works as intended between the viewer and the AI Model Service. After uploading a DICOM image with proper tags to ORTHANC and opening the study from the study list, it triggered the AI Model Service. The results from the AI Model Service then appeared on the viewer without any errors.

Moving forward into next week, I will align on my Merge Requests (MRs), make adjustments based on my mentor's feedback, and then proceed with modeling.

Comments

Post a Comment

Popular posts from this blog

Google Summer of Code 2023: AI Model Service Auto Selection by Modality - Final Report

Image
Introduction GSoC is one amazing program that lets students and professionals get into the world of open source, contribute code and also learn back from the program. I definitely learnt a lot by contributing to my mentoring organization LibreHealth through GSoC'23. I now understand the value of open source and will become a lifelong contributor wherever and whenever I can. LibreHealth is an umbrella organization where multiple health information technology projects collaborate. I was actively involved in contributing code to their LibreHealth Radiology application  through the project "AI Model Service Auto Selection by Modality". My contributions to the application were more involved into implementing AI models in the application and also building modules that do automatic model selection, example, when a Radiologist opens up a Mammography image the application automatically triggers the Mammography AI model to produce inferences on the image. Objectives The project des
Read more

GSoC Week 1

This week I was researching AI models to implement for the different modalities. More specifically, Mammography. I also finally had the opportunity to talk to my mentor on Thursday (June 1st, 2023). You can find the recorded meeting link  here . Building a Breast Cancer Classification Model from Scratch At the start of the week, I attempted to implement a VGG16 classifier to distinguish between malignant and benign calcifications in a mammogram. Using PyTorch, I developed the model and achieved a validation accuracy of approximately 70%. The source code can be accessed  here . This model and code serve as a starting point for implementing a more complex model mentioned in this  paper . Briefly, the paper discusses building a model solely on the region of interest (ROI) of the calcification/tumor and adding layers and training on the entire images. This method has significantly reduced the training loss and, consequently, improved overall accuracy. Mentor Discussion and Change in Direct
Read more

My First GSoC - LibreHealth Community Bonding Period

GSoC Acceptance I was amazed to know that I got accepted into Google Summer of Code 2023. This is my first time applying and getting into GSoC. I remember the time when I sat at the library for hours, working on the preliminary tasks, figuring out solutions, and writing an extensive proposal. Reiterating through the solution and proposal back and forth, I watched lots of YouTube videos and read blogs providing tips to draft a good proposal. All the efforts paid off when I received the email of acceptance. The Org & Project I drafted a proposal to the LibreHealth Organization. LibreHealth is collaborative community for free & open source software projects in Health IT, and is a member project of Software Freedom Conservancy. It is amazing to just think about how contributors from all over the world come together and build things that impact the world positively. This idea is what brought me into open source. Before jumping into the project details, here are a few things about m
Read more