Healthcare
Diabetes Readmission Risk Analysis
A pre-discharge readmission score built around patient-level validation, discharge-disposition leakage and the care team's real follow-up capacity.
View projectIndustry · 04
Analytics for care providers and health insurers: readmission risk, claims and fraud review, operational reporting and privacy-preserving AI.

In the health insurance industry, there is a growing need for efficient management of patient data, streamlined claims processing, and robust fraud detection measures.
We use predictive analytics to assess risk and support care and claims decisions. Patient data is treated as sensitive by default: minimum necessary fields, de-identification where possible, role-based access and a documented retention policy. We work within your privacy and regulatory constraints rather than around them.
We also apply AI carefully: retrieval assistants over clinical policy, document processing for claims, and predictive models that support — never replace — clinical judgement.
4 projects
Healthcare
A pre-discharge readmission score built around patient-level validation, discharge-disposition leakage and the care team's real follow-up capacity.
View projectHealthcare
How a clinical research group analysed a small cirrhosis cohort with time-to-event methods — censoring, competing risks and an observational treatment comparison reported as what it is.
View projectHealthcare
A small clinical dataset used to show what a mortality score can and cannot support — time-to-event framing, calibration, class balance and the limits of subgroup analysis at 299 patients.
View projectHealthcare
How a care provider turned readmission risk into a discharge-time worklist clinicians would actually use — cohort rules agreed up front, overrides logged and privacy designed into the workflow.
View projectFocus on your business; let us back you up with data support