See the announcement for a post-doc position at OSU's Imageomics Institute. 
Learn more about the Imageomics Institute and its team: 
https://imageomics.osu.edu/team 
<https://urldefense.com/v3/__https://imageomics.osu.edu/team__;!!K-Hz7m0Vt54!mSsoxcd_DQxkFRFYDut0jfvQylNLLpMhAjk701dfvXbNp8ED5ppBgyP3RwM-ODh3ZYo8tFSOdFyrTHAWmSGJB2Ja0ZDSzWc$>
 

Informal inquiries are welcome; feel free to email me directly.

Best regards,

*A. Murat Maga, PhD*
Professor
Division of Craniofacial Medicine
Dept of Pediatrics
University of Washington
&
Principal Investigator
Center for Developmental Biology and Regenerative Medicine
Seattle Children's Research Institute
https://faculty.uw.edu/maga 


Postdoctoral position in 3D Artificial Intelligence for Biological Imaging
Based at Ohio State University, the postdoc will be jointly supervised by 
Dr Tanya Berger-Wolf and Murat Maga, as part of the Institute's 3D 
Convergence Working Group. Please share this with anyone who might be 
interested.
https://osu.wd1.myworkdayjobs.com/OSUCareers/job/Columbus-Campus/Post-Doctoral-Scholar_R159228
 
<https://urldefense.com/v3/__https://osu.wd1.myworkdayjobs.com/OSUCareers/job/Columbus-Campus/Post-Doctoral-Scholar_R159228__;!!K-Hz7m0Vt54!mSsoxcd_DQxkFRFYDut0jfvQylNLLpMhAjk701dfvXbNp8ED5ppBgyP3RwM-ODh3ZYo8tFSOdFyrTHAWmSGJB2JaEDO7s2g$>
 

The project: Biological imaging now produces tomographic volumes, surface 
meshes, point clouds, and geometric measurements across hundreds to 
thousands of specimens, far more than manual analysis can handle. The 
postdoc will develop AI methods that turn these data into quantitative 
biological measurements and apply them to open questions in organismal and 
evolutionary biology.

The core of the work is 3D segmentation and adapting foundation models to 
volumetric biological data, including:
- Evaluating and adapting promptable and automated segmentation models for 
volumetric imaging, including models trained on very different data
- Atlas- or template-based methods for initializing and propagating 
segmentations across specimens, benchmarked against expert annotation
- Evaluation protocols that honestly report where and why methods fail

The candidate will also contribute to at least one of:
- 3D shape analysis and geometric ML: learning on meshes, point clouds, and 
landmarks; automated landmarking; correspondence across specimens; scaling 
geometric morphometrics to large collections
- Multimodal and generative approaches: linking 3D structure to text and 
images, retrieval and description of 3D specimens, generative models of 
biological shape

Required qualifications:
- PhD (completed or expected before start) in computer science, biomedical 
engineering, applied math, statistics, a quantitative biological 
discipline, or related field
- Research experience in machine learning or computer vision (publications, 
preprints, or public code)
- Strong Python skills and experience with PyTorch or a similar framework
- Experience with 3D data (volumes, meshes, point clouds, or geometric 
representations)

Preferred: volumetric segmentation experience (annotation, class imbalance, 
small training sets); foundation model adaptation (fine-tuning, prompting, 
PEFT); geometric deep learning or statistical shape modeling; 
vision-language/multimodal models; biological or biomedical imaging; 
HPC/cloud computing; open-source scientific software contributions.

Details: One-year, full-time term position, available immediately. Review 
of applications begins immediately and continues until filled.

To apply: Submit a cover letter, CV, a statement of research skills and 
interests, and contact information for three referees (with your 
relationship to each) here:

   - 
   
https://osu.wd1.myworkdayjobs.com/OSUCareers/job/Columbus-Campus/Post-Doctoral-Scholar_R159228
 
   
<https://urldefense.com/v3/__https://osu.wd1.myworkdayjobs.com/OSUCareers/job/Columbus-Campus/Post-Doctoral-Scholar_R159228__;!!K-Hz7m0Vt54!mSsoxcd_DQxkFRFYDut0jfvQylNLLpMhAjk701dfvXbNp8ED5ppBgyP3RwM-ODh3ZYo8tFSOdFyrTHAWmSGJB2JaEDO7s2g$>
    



-- 
You received this message because you are subscribed to the Google Groups 
"Morphmet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/morphmet2/66b13561-c957-41c6-9b7c-23cb90431b61n%40googlegroups.com.

Reply via email to