The increased availability of highly effective tools and image data sets has made AI use for medical imaging research more accessible than ever before.
In radiology, artificial intelligence (AI) is the engineering of computerized systems that can perform tasks typically done with human intelligent behavior such as acquiring, reconstructing, analyzing, and/or interpreting medical images. Machine learning (ML) algorithms are a subset of artificial intelligence methods that computerized systems use to solve problems by recognizing patterns in the data. AI has been developed to improve radiology workflows and assist radiologists with tasks such as lesion detection and quantification of medical images. Those exploring artificial intelligence and machine learning in medical imaging research require software tools to assist in algorithm development and image data to train and test algorithms.
Many tools and resources exist that can help develop AI algorithms for medical imaging such as:
- Figma, which allows multiple team members to create wireframes and mockups of proposed designs;
- TensorFlow, which provides a collection of workflows to develop and train models using Python, JavaScript, or Swift;
- Tribuo, which provides a Java Machine learning library to code algorithms;
- ML.NET, a software machine learning library for the C# and F# programming languages;
- PyTorch, a machine learning framework with a Python interface and C++ interface; and
- Keras, a library that provides a Python interface for artificial neural networks. These tools can assist developers to build and then later deploy their machine learning applications based on their preferences and expertise.
Most developing AI algorithms cannot directly access a picture archiving and communication system (PACS) environment. Traditionally, there has been a perception that a lack of adequate medical image data exists due to small sample sizes and a lack of geographic diversity. However, some large datasets now have high quality images and annotations. These datasets have gone through DICOM de-identification to meet the U.S. Health Insurance Portability and Accountability Act (HIPAA) requirements. These datasets also include image labels, which are annotations performed by radiologists. Some AI algorithms need annotations, which serve as ground truth, for medical image classification based on a supervised learning approach. The table below offers a list of accessible medical imaging datasets. These datasets include those available from Kaggle, the Cancer Imaging Archive (TCIA), the National Institutes of Health (NIH), and Stanford University’s Center for Artificial Intelligence in Medicine.
Once downloaded, the cloud can store these datasets to improve developers’ ability to share and provide data back abilities. Amazon Simple Storage Service (Amazon S3) can perform cloud-based storage of datasets. MongoDB loads files such as JSON files with annotations linked to the image data sets. Tools like Django and Flask can build web apps for demonstration once machine learning algorithms are trained with medical imaging datasets,. Additionally, AWS SageMaker demonstrates the purposes of machine learning models.
Familiarity with tools available to assist with AI algorithm development and availability of image data to test the algorithms is necessary for implementing successful AI algorithms in medical imaging research. With improved awareness of the numerous tools and data sources available to AI researchers today, we can increase AI participation in medical imaging research and ultimately accelerate progress being made in AI’s medical imaging transformation.
If you’re ready to explore the potential of AI to solve your organization’s medical imaging research needs, Ellumen’s AI experts are here to help. Keep an eye out for the next part of Ellumen’s blog series exploring topics related to AI innovation in medical imaging.
Meet Ellumen’s AI Experts
Are you looking to utilize AI for your next technology or integrate AI into your current systems? Feel free to reach out to us.
Iyanuoluwa Odebode is Ellumen’s artificial intelligence/machine learning expert, supporting the Innovation Lab project by identifying and researching new & emerging technology. He also serves as an adjunct professor in cybersecurity at University of Maryland Baltimore County. Prior to his work at Ellumen, Iyanuoluwa built an algorithm for repurposing old drugs for new use using the DReiM methodology. He has been published in ResearchGate and IEEE for his research on machine learning. Iyanuoluwa completed his master’s in bioinformatics at Morgan State University and his Ph.D. in information systems (machine learning/AI) at University of Maryland Baltimore County.
Todd R. McCollough is a Software Engineer for Ellumen. He works primarily with the CVIX/VIX Services which support image viewers and applications to provide the ability for users to query, retrieve and manipulate VA and DoD medical images and image artifacts. Todd is a co-inventor on several issued U.S. patents and is passionate about discovering novel ways to image patients and improve patient care. Todd received his bachelor’s and master’s degrees in biomedical engineering from Northwestern University.