AI Clinical Outcome Prediction Researcher

Lamion Solutions

Hybrid · $200k+ / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

We serve 28 health systems and academic medical centers. Our NLP systems already extract structured information from clinical notes. The next capability our clients want — and are willing to pay significantly for — is predictive: given a patient's clinical record at time of admission, what are their likely outcomes? ICU admission probability. 30-day readmission risk. Length of stay distribution. Complications conditional on proposed treatment path. Clinical outcome prediction is harder than most ML applications because the training data reflects clinical decisions, not random assignment. Patients who receive intensive early intervention are systematically different from those who don't, and naive models learn those correlations without disentangling intervention effects from risk factors. You need causal reasoning, not just pattern matching. You'll lead the outcome prediction research program: designing models that account for treatment selection bias, validating against held-out hospital systems, and implementing the fairness testing that ensures models perform equitably across patient demographics. You'll publish your methods — our partners expect research-grade rigor, not black-box systems. The role is hybrid: 2 days per week in our Boston office for clinical collaborator meetings.

Responsibilities

Design clinical outcome prediction models accounting for treatment selection bias
Implement causal inference techniques to separate risk factors from intervention effects
Validate models across held-out hospital systems with different patient populations
Develop fairness testing ensuring equitable performance across demographic groups
Publish methodology in clinical informatics or ML-for-health venues
Present findings to clinical partners and hospital leadership

Requirements

7+ years in ML research with clinical or health data science experience
Deep expertise in causal inference and confounding adjustment in observational data
Strong Python and statistical computing skills with large EHR dataset experience
Publication record in clinical ML, bioinformatics, or related health informatics venues
Ability to collaborate with physicians and clinical informaticists