Reducing Hepatitis C diagnostic disparities with a point of care assay for HCV antigen detection
Data files
Mar 06, 2025 version files 8.03 MB
-
design.zip
8.03 MB
-
README.md
4.61 KB
Abstract
Viral hepatitis continues to be a significant global health issue, with chronic hepatitis B (HBV) and hepatitis C (HCV) responsible for approximately 1 million deaths each year, primarily due to complications such as liver cancer and cirrhosis. Annually, more than 1.5 million individuals contract HCV, with vulnerable populations, including American Indians and Alaska Natives (AI/AN), being disproportionately affected. Although direct-acting antivirals (DAAs) have proven highly effective, the timely and accurate diagnosis of HCV remains a major challenge, particularly in settings with limited resources. The current two-step HCV testing approach is both costly and time-intensive, often resulting in patient loss before appropriate care is administered. Point-of-care (POC) HCV antigen (Ag) testing presents a viable alternative, offering the potential for early detection, even during the acute phase of infection. However, there is currently no FDA-approved POC HCV Ag test that meets the required sensitivity and specificity for detecting low viral loads.
Design files
All the designs for the 3d printed objects and laser cut objects used in the manuscript.
The file “3D printing parts.f3d” contains all the 3D designs of the 3D printed parts used in the manuscript, and can be opened with Autocad Fusion 360.
The file “Layer-1.ai” contains the 2D designs of laser cutting path of the 1st layer of PMMA sheet of the microfluidic chip used in the manuscript, and can be opened with Adobe Illustrator.
The file “Layer-2.ai” contains the 2D designs of laser cutting path of the 2nd layer of PMMA sheet of the microfluidic chip used in the manuscript, and can be opened with Adobe Illustrator.
The file “Layer-3.ai” contains the 2D designs of laser cutting path of the 3rd layer of PMMA sheet of the microfluidic chip used in the manuscript, and can be opened with Adobe Illustrator.
The file “Layer-4.ai” contains the 2D designs of laser cutting path of the 4th layer of PMMA sheet of the microfluidic chip used in the manuscript, and can be opened with Adobe Illustrator.
The file “Layer-5.ai” contains the 2D designs of laser cutting path of the 5th layer of PMMA sheet of the microfluidic chip used in the manuscript, and can be opened with Adobe Illustrator.
Requirements
Autocad Fusion 360
Adobe Illustrator
Microfluidic code
Running codes for the microfluidic device.
Requirements
Arduino IDE 2.3.4
SPyDERMAN
Smartphone-based Pathogen detection resource multiplier using adversarial networks (SPyDERMAN)
This is the Pytorch implementation for our paper Mobile Health (mHealth) Viral Diagnostics Enabled with Adaptive Adversarial Learning.
Abstract:
Deep-learning (DL)-based image processing has potential to revolutionize the use of smartphones in mobile health (mHealth) diagnostics of infectious diseases. However, the high variability in cellphone image data acquisition and the common need for large amounts of specialist-annotated images for traditional DL model training may preclude generalizability of smartphone-based diagnostics. Here, we employed adversarial neural networks with conditioning to develop an easily reconfigurable virus diagnostic platform that leverages a dataset of smartphone-taken microfluidic chip photos to rapidly generate image classifiers for different target pathogens on-demand. Adversarial learning was also used to augment this real image dataset by generating 16,000 realistic synthetic microchip images, through style generative adversarial networks (StyleGAN). We used this platform, termed smartphone-based pathogen detection resource multiplier using adversarial networks (SPyDERMAN), to accurately detect different intact viruses in clinical samples and to detect viral nucleic acids through integration with CRISPR diagnostics. We evaluated the performance of the system in detecting five different virus targets using 179 patient samples. The generalizability of the system was confirmed by rapid reconfiguration to detect SARS-CoV-2 antigens in nasal swab samples (n = 62) with 100% accuracy. Overall, the SPyDERMAN system may contribute to epidemic preparedness strategies by providing a platform for smartphone-based diagnostics that can be adapted to a given emerging viral agent within days of work.
Requirements
- Python 3.5
- Pytorch 1.4.0
- PyYAML 5.3.1
- scikit-image 0.14.0
- scikit-learn 0.20.0
- SciPy 1.1.0
- opencv-python 4.2.0.34
- Matplotlib 3.0.0
- NumPy 1.15.2
- TensorFlow 1.10.0 (For TensorBoard)
SPyDERMAN
Usage
Generating synthetic data library:
Run python scripts/stgan.py
Adversarial training with data library and target virus:
Run ` python scripts/train_image_.py –dset “target viral dataset location”`
example: ` python scripts/train_image_.py –dset “COVID19”`
Citing
We encourage citation of our paper if you use our code in your research:
@article{shokr2020mobile,
title={Mobile Health (mHealth) Viral Diagnostics Enabled with Adaptive Adversarial Learning},
author={Shokr, Ahmed and Pacheco, Luis GC and Thirumalaraju, Prudhvi and Kanakasabapathy, Manoj Kumar and Gandhi, Jahnavi and Kartik, Deeksha and Silva, Filipe SR and Erdogmus, Eda and Kandula, Hemanth and Luo, Shenglin and others},
journal={ACS nano},
year={2020},
publisher={ACS Publications}
}
Contact
If you have any questions, please contact us via hshafiee[at]bwh.harvard.edu.