GSoC Week 7

This week I was able to discuss with my mentor on  my progress and the merge requests. I took him through the changes I made. My mentor suggested a few minor updates to make and there were methods for which the GET method were not setup and we also questioned what their functionality would be.

Updates to the Bounding Box Endpoint

The /bounding-box endpoint was necessarily a POST method where payload consisting of the image and its relevant information is sent to perform model inference and return them back. The GET method which was previously defined returned all models available in the application. It was redundant to the /models method. We decided to update the /bounding-box GET method to return only the models available under it and their versions. For this task I reorganized the folder structure from models/cheXnet to models/bounding_box/cheXnet. This update opens up a lot of flexibility in terms of integrating models.

In future we can have more models under /models/bounding_box and calling the GET method of the /bounding-box would return all models under it. Again if we need to integrate just a classifier based model we can create models/classifier in our folder structure and integrate models under it.

Challenges with Test Cases


Once the updates were made there were test cases that started failing. I had to check them and update the test cases. Since the functionality of the endpoints were changing they needed to be updated. There is still one particular test case in the lh-radiology application that I am not able to figure out. It provides different results compared to my local system and the gitlab CI/CD.

I created a new test case for the \model-selection method but it was failing. In the /model-selection method we are capturing the DICOM tags of the image, finding the modality and redirecting to the appropriate model. After some reading I figured redirecting within flask and trying to do unit tests using pytest would fail but more importantly I read redirecting within a flask endpoint is not right since it increase complexity and reduces performance. I think now it is better to just call the method by importing them instead of calling its endpoint. I will need to discuss this with my Mentor before moving forward.

Rebasing Commits 

I made some multiple comments which I rebased to single comments for readability.

Upcoming Tasks

I will work on refactoring the model selection code to call the methods from imports rather than endpoints. Then I will start working on implementing a Mammography tumor classification model. I believe more or less the model selection hook is complete.


The cheXnet model after discussion with my mentor is good to go and now the outputs are fine now,



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