Enteritis scoring on distal intestine slides are critical measures for industry study reporting on the effects of alternative feed formulation and dietary ingredients. Limited progress has been made in formalizing these measurements through grading rubrics and rules of thumb (Uran 2008). Current scoring relies on manual grading by a professional histopathologist against a prescribed ordinal scale on several anatomical metrics. Due to the slow, expensive, and inconsistent nature of this human-based approach, there is still significant room for improvement in the cost, accessibility, and robustness of enteritis grading. This proposal aims to train a machine-learning model to rapidly, accessibly, cheaply, and robustly quantify distal enteritis in histology slide images in a bias-free and reproducible manner (Guan, 2022). This technology may be applied to every future project involving distal enteritis evaluation. This project addresses priority 2.1 of the SAA RFP: Understanding gastrointestinal barriers to soy inclusion: specifically, enteritis. We propose that a successfully trained model may achieve enteritis grading at or above the level of a board-certified histopathologist, on the order of milliseconds, at zero cost, in a freely available format which is usable by anyone with access to a gpu-enabled laptop.

(1) Crowdsource 800-1000 commercial salmonid (Atlantic salmon and rainbow trout; approx. equal proportions) distal intestinal slides from multiple industry and research partner laboratories, ranging along the continuum of clinical enteritis from negative (zero inflammation) controls to severe distal enteritis samples, including representatives from all intermediate stages. Efforts will be made to include divergent sources, including domestic and international collaborators (i.e., USA, Norway, Chile, etc.), to improve the variability within the initial training data. Training data variability is expected to increase the robustness and generalizability of the model.
(2) Digitize the obtained slides using a professional-grade slide scanner and assign binary classifications and Uran scores to each digitized slide utilizing two to three independent histopathologists.
(3) Train the ResNet AI model as a binary classifier as proof of concept on these digitized images, to judge each slide as a binary enteritis-positive or enteritis-negative.
(4) Train the model to assign each slide an ordinal grade on each standard anatomical metric, according to the Uran Scale (He, 2016; McCombe, 2021).

(1) A freely accessible, easily transferable lightweight machine-learning program known as a ‘ResNet’ that can grade distal intestine slide images on the Uran Scale and return data on the order of milliseconds per slide at zero cost.
(2) A tool to drastically reduce the cost and turn-around time of enteritis data for every scientific investigation into soy-based diets.
(3) Demonstrate the economic value of machine-learning and artificial intelligence-assisted aquaculture tools to farmers, stakeholders, and economic partners (Sveen 2021; Vo 2021).

Updated June 12, 2024:
Executive Summary
The project, aimed at developing a machine-learning model to score distal intestinal histology slides for soybean meal-induced enteritis, was initial intended to officially commence on April 1, 2024; however, due to administrative delays, the contract was not signed until May 20, 2024 and a formal start date of May 1, 2024 was request. Despite these delays, significant progress was made in these initial months of the project, with a focus on algorithm development and initial model construction.

Q1 Project Activities and Achievements

1. Meetings and Planning:
Dr. Jacob Bledsoe (PI) and Dr. Nathan Redman (coPI) held multiple sessions to outline the construction of the ResNet algorithm and the sourcing of necessary histology slides. A pivotal 2-hour meeting involving the Project PI, Dr. Nathan Redman, and machine learning expert Anita Juhong occurred on May 25th. This meeting focused on determining the optimal parameters for constructing the ResNes analysis network, crucial for the project's success.

2. Development and Scripting of Neural Network:
The team, led by Dr. Nathan Redman, successfully developed and scripted version 1.1 of a ResNet-style architecture neural network utilizing the PyTorch framework. This version includes advanced features such as cross-entropy loss and a resource management CUDA module, ensuring efficient processing and management of computing resources.

3. Image Preprocessing Development:
Dr. Nathan Redman has begun initial scripting and development of the image preprocessing module, crucial for preparing raw histology slide data for neural network analysis. Utilizing Python Image Library and OpenCV, this module ensures that the raw image data are appropriately formatted to be optimized for input into the neural network.

4. Collaboration Changes:
Dr. Liam Neiswanger-Broughton of Washington State University had initially agreed to participate as a collaborating histopathologist, but has had to step away from the project. Discussions are currently underway with other potential histopathologists to assist with the manual ground-truth scoring of histological slides, a critical component for training our model.

Challenges and Adjustments
The delayed formal signing of the project contract posed initial administrative challenges; however, the team adapted quickly, ensuring that project milestones remained on track. The unexpected withdrawal of Dr. Broughton-Neiswanger necessitated a search for additional expertise, which is currently being addressed to minimize impact on the project timeline.

Next Steps
For the upcoming quarter, the focus will be on:
1. Finalizing collaborations with new histopathologists.
2. Beginning the collection and preprocessing of histology slides as per the project timeline.
3. Further refining and testing of the neural network model to ensure robustness and accuracy.

View uploaded report

Updated September 16, 2024:
Executive Summary

This quarter has seen advancements in both technical development and team dynamics. Despite the unexpected departure of Dr. Nathan Redman, our lead research scientist, the project continues to progress well with the completion of the ResNet algorithm scripting and the strengthening of our collaborative network through the addition of Dr. Salvatore Frasca, a highly regarded histopathologist.

Q2 Project Activities and Achievements

1. Staffing Updates:
-Dr. Nathan Redman, our lead research scientist/postdoc, has taken a new position effective September 13th. A broad search for his replacement is already underway, with advertisements placed at key conferences such as Mucosal Health in Aquaculture 2024 and WAS 2024 Copenhagen, as well as on university-related job boards and social media accounts. We are currently reviewing new applicants.
2. Completion of Machine Learning Framework:
-The machine learning algorithm, ResNet, has been fully scripted, and the computational framework is complete. This ensures that we have a reliable system for processing and analyzing digitized histology slides.
3. Collaboration and Crowdsourcing:
-Dr. Salvatore Frasca, a skilled and well-respected histopathologist, has agreed to participate in the project. His involvement is crucial for the scoring of digitized slides, particularly following the departure of our initial histopathologist.
-We have had success in crowdsourcing slides, particularly following our advertisement at Mucosal Health Aquaculture 2024 in Porto, Portugal. We have received contact from multiple laboratories interested in sharing slides from Atlantic salmon and trout, which enhances the diversity and volume of our data set.

Challenges and Adjustments

The departure of Dr. Nathan Redman posed a significant challenge this quarter, necessitating a swift and broad-reaching search for a qualified replacement to ensure continuity in our research activities. Additionally, securing a new collaborating histopathologist was successfully accomplished with the involvement of Dr. Frasca, mitigating potential delays in our timeline.

Next Steps

In the next quarter, we will focus on:
• Finalizing the recruitment of a new lead research scientist.
• Scoring of more histology slides with Dr. Frasca.
• Continuing to collect and process histology slides from international collaborators.
• Further testing and refining of the neural network model to ensure its readiness for practical application.
• Optimizing the model to include a pre-training quality assurance/quality control (QA/QC) and image subsampling step. This will ensure that images are not biased and allow the model to effectively subsample each image, enhancing the robustness and accuracy of our analysis.
• Iterative optimization of the ResNet parameters will continue as the model is further trained on ground-truthed histopathologist scored slides.

View uploaded report

Updated December 16, 2024:

View uploaded report

View uploaded report 2

Updated March 21, 2025:
Summary
In this quarter, administrative progress has been made in addressing previous project delays. A No-Cost Extension (NCE) request was submitted on February 11, 2025 and approved by the Soy Aquaculture Alliance (SAA) on March 18, 2025, extending the project termination date to December 31, 2025. This extension allows for completion of key project objectives, particularly the collection of a sufficiently large and diverse set of histological slides for training the machine learning model. Additionally, a research associate has been hired onto the project to replace the research scientist that left the project in August 2024. The new hire has started working on expanding collaborations to acquire additional samples.
Q1 2025 Project Activities and Achievements
1. No-Cost Extension Approval:
• The project’s NCE request was approved by SAA and is currently being processed by the University of Idaho’s Office of Sponsored Projects.
• The extension was necessitated by delays in hiring a research scientist (who left in August 2024), which subsequently delayed the acquisition of a sufficient number of histological slides from salmonid studies containing both healthy and enteritis-positive individuals.
2. Hiring of Research Associate and Expanded Slide Sourcing Efforts:
• A research associate was hired and has allocated 36% FTE to this project.
• They have begun contacting additional external labs and collaborators to source more histological slides.
• Efforts are being made to ensure the dataset includes a broad representation of salmonid intestinal samples, particularly those capturing a range of enteritis severity.
3. Machine Learning Model Preparedness:
• The slide processing and machine learning model frameworks are complete and prepared for training.
• We are waiting to secure a sufficiently large dataset of slide images to ensure robust training of the model.

Challenges and Adjustments
• Slide Sourcing Delays: While the machine learning framework is ready, the project still requires a larger, high-quality dataset of histology slides for training.
• Hiring Delays: The delayed hiring of a research scientist set back progress on sourcing efforts, though the recent hiring of a research associate has helped mitigate these issues.

Next Steps
• Continue external collaboration efforts to acquire additional histology slides, with a focus on ensuring both healthy and enteritis-positive samples.
• Finalize processing of the NCE through the Office of Sponsored Projects.
• Begin initial training runs of the machine learning model with available histology slide data.
• Evaluate slide dataset gaps and determine if additional sourcing efforts are required before full-scale training can commence.

View uploaded report

Wet lab tests and farm trials that test effects of any parameter (e.g., breeding programs, supplements, probiotics, antinutritional factors) on distal enteritis are critical for increasing levels of soy protein replacement in fish feed (Booman, 2018). Professional grading of slides costs tens of thousands of dollars and may vary in quality and time requirements. This approach makes the grading process instantaneous, reliable above 95%, and cost-free, removing a hurdle to increase soy utilization in aquaculture.