The Center for Clinical Data Science is focused on the full lifecycle of product development for machine learning enabled clinical applications, from early stage R&D efforts through to clinical validation and commercialization via our channel partners.
In addition to internally developed applications, the CCDS works with clinicians and researchers on promising early stage research and development efforts. We provide a broad set of services and support for R&D activities, including computational support, data pipeline tools and infrastructure, as well as data science and software development expertise.
The CCDS is currently providing early stage R&D support to more than 25 projects across multiple specialties at both Mass General and Brigham and Women’s Hospitals.
In addition to early stage efforts, the CCDS is supporting the development and validation of several AI enabled product prototypes.
Frequently Asked Questions (FAQ)
+ What is the CCDS Mission, where are we located, and with whom do we work?
The Center for Clinical Data Science is an interdisciplinary team of data scientists, clinicians, technologists, and developers dedicated to the goal of translating advances in data science and machine learning for the betterment of healthcare worldwide. We focus on the entire translational pipeline – from model conceptualization to clinical validation in the Partners Healthcare System. We are located in Boston but are fortunate to work with collaborators worldwide, encompassing research institutions, medical centers, and industry. We acknowledge we cannot achieve this goal alone, so we work with dedicated experts across a variety of skill sets, disciplines, and locations.
+ How are we different from other data science centers in the broader community?
- As part of Partners Healthcare, we have an intimate understanding of patient and clinician end-user needs and healthcare workflows. As such, we are able to identify high value opportunities with the end-goal of translating solutions into clinical practice.
- We have access to unparalleled research and data science expertise internally coupled with clinical domain expertise through the Massachusetts General Hospital and Brigham and Women’s Hospital.
- We have access to significant stores of clinical data and computing backend.
- With our focus on clinical translation and product development, we combine these critical elements to create solutions that will be adopted to transform care.
+ What types of projects do we currently support?
We work on a wide variety of projects, but the two most common types are: 1) projects that address a clearly articulated medical problem or clinical workflow pain point and have the support of a “champion” (in clinical medicine or elsewhere within the healthcare enterprise) that could be addressed utilizing machine learning, and 2) novel transformative ideas that could benefit the medical and/or machine learning disciplines by leveraging CCDS data and resources. While many of our initial projects are focused on elements of diagnostic medicine, including radiologic images, we also entertain projects within clinical informatics as well as healthcare enterprise operations.
+ With what types of collaborators do we typically work?
We believe that a broad collection of stakeholders – including researchers, healthcare providers and other professionals, and data scientists – are needed to translate advances in machine learning into clinical practice. We work with people across the Charles River, throughout the United States, and across the Atlantic Ocean. We also work with major industry partners as well as startups to support these projects, and to translate them into clinical practice. We aim to quickly provide these solutions to the Partners Healthcare enterprise and beyond.
+ What resources are available to support researchers’ and clinicians’ projects?
- We support researchers and clinicians through a variety of means, and our level and type of support varies depending on the needs of our collaborators. We often provide a mix of human capital (machine learning expertise, software development, product development, etc.), infrastructure, and computational support.
For project execution, we provide a host of capabilities and resources needed to translate a project from concept to clinical use. These include:
- Computational infrastructure (GPU, storage, dedicated clusters, etc.)
- Machine learning and data science / engineering expertise
- Data access and annotation
- Software development
- Clinical testing and workflow integration
- Product development
- Access to industry partnerships for translation
- We do not see ourselves as a core lab or fee-for-service entity. Rather, we see ourselves as a collaborative partner in research, development, and translational efforts to advance the cause and impact of machine learning in healthcare.
+ What is our intake mechanism for projects and partners?
- We want interfacing with the center to be easy – to contact us, you can send us an e-mail at firstname.lastname@example.org. To make sure that we use your time efficiently, we ask that you describe your project idea in abstract form highlighting the clinical or research question, proposed solutions, and the direct needs that you foresee. We have created an outline as a guide that we have found useful in thinking through our own projects; please use this intake form.
- You can also sign up for our mailing list, through which we will update you on upcoming events, such as our regular journal club. These events also serve as an informal forum to discuss potential projects. Please join us!
- Additionally, if you are on the medical side of things, we host biannual “Clinicians in Data Science” meetings at MGH and BWH where potential clinical collaborators have the opportunity to learn about data science in healthcare, meet each other, and get a chance to interact with us here at the center. If you are a startup or new technology entity looking to engage with us, please tell us about your company and your idea.
+ What is the CCDS Policy on IP sharing/rights?
We follow the Partners policy on Intellectual Property (IP) rights and sharing. We use several frameworks for this IP, depending on who you are (researcher or industry), and the degree of collaborative involvement. While details fall outside the scope of a FAQ, we typically discuss IP upon initiation of a project together.