The Musculoskeletal Digital Innovation and Informatics (MDI²) Program in the Department of Orthopedics and Sports Medicine at Boston Children's Hospital and Harvard Medical School has an immediate opening for an outstanding, highly motivated Postdoctoral Research Fellow to work on several image processing and modeling projects to assist with clinical decision making in patients with musculoskeletal joint disorders. The fellow will join a multidisciplinary team of biomedical and mechanical engineers, orthopedic surgeons, and computer and data scientists developing tools that analyze medical images (e.g., CT and MRI) to support the diagnosis and treatment of joint disorders.
The central theme of the work is the development of automated, end-to-end tools for comprehensive assessment of joint anatomy and function in three dimensions: reconstructing patient-specific bone models from routine clinical imaging, quantifying morphology to localize the type and extent of anatomical abnormality, simulating joint articulation across ranges of motion representative of daily and athletic activity, and interpreting the results against normative reference data. These tools are intended to run without manual intervention, at cohort scale, and to be reliable enough that clinicians can act on their output, which places a premium on methods that behave predictably across the full spectrum of anatomy encountered in practice, from normal through deformity.
This fellowship opportunity is primarily focused on translational research with immediate impact on clinical care of patients with a range of spinal conditions, through publication of novel findings and development of computational tools used by clinicians. The postdoctoral fellow will have joint appointments at Boston Children's Hospital and Harvard Medical School and will benefit from access to the largest medical R&D community in the world (Longwood/Boston), along with numerous career development resources offered by Harvard University and its affiliated hospitals and research centers.
Required skills and experience
- PhD in mechanical or biomechanical engineering or a related field.
- Strong foundation in analytical and computational geometry and in 3D rigid-body kinematics. For example, definition and construction of coordinate systems, spatial transformations and transformation matrices, and the properties and limitations of common rotation representations, together with an understanding of why anatomically referenced frames are a prerequisite for meaningful comparison between subjects.
- Expertise with Python.
- Experience working with medical images and/or 3D geometric data (e.g., volumetric images, surface meshes, point clouds).
- Demonstrated independent research output (e.g., first-author publications) and the ability to carry a loosely specified problem through to a result.
- Excellent communication skills in English.
Desired skills and experience
- Experience developing algorithms that perform reliably across a wide range of normal and pathological anatomies.
- Computer simulation of joint motion or multibody kinematics; finite element analysis.
- Basic knowledge of human musculoskeletal anatomy, biomechanics, and pathology.
- Sound software practice (e.g., version control, testing, reproducible pipelines) and experience processing data at scale (e.g., batch or high-performance computing).
- Statistical methods for measurement validation and reliability.
- Registration and correspondence methods (e.g., rigid and non-rigid registration, statistical shape modeling, landmark propagation).
- Experience with generative AI and large vision models.
- Experience generating 3D anatomical models from medical images (e.g., segmentation, surface and mesh processing).
Appointment
This is a full-time position renewed annually based on the applicant's performance.
How to apply
Send the following materials to MDI2.Program@childrens.harvard.edu:
- Curriculum vitae.
- A copy of a technical paper.
- A paragraph stating your goals.
- Contact information for three references.