Precision Medicine Analytics

WE SUPPORT YOUR PERSONALIZED
HEALTH QUESTIONS USING AI

Precision Medicine Analytics

WE SUPPORT YOUR PERSONALIZED
HEALTH QUESTIONS USING AI

Our Precision Medicine Analytics Services

Our multi-disciplinary team generates precise results for identifying patient subgroups, predicting treatment effectiveness, and enabling personalized diagnoses.

Targeted Subgroup Analysis

Identify and reach distinct patient subgroups by clinical severity, access, and persistence, ensuring your treatments are effective for those who need them most.

Individualized Effectiveness Predictions

Predict which care for whom. Our AI-driven insights forecast treatment effectiveness at the individual level to support precision medicine and optimize patient outcomes.

Personalized Diagnosis

Improve diagnosis and early diagnosis. Our AI-driven insights predict diagnosis and comorbidities at the individual level to support early patient care.

From Real-World Data to Certified Clinical Impact

Case Study: PHREND® – Real-World Precision for MS Treatment Decisions

From Real-World Data to Certified Clinical Impact

The Challenge of Personalized Treatment in MS. Treatment decisions in relapsing-remitting multiple sclerosis (RRMS) are highly individualized. Each patient presents with a different clinical profile, disease progression, and personal preference. Despite the growing number of therapy options, clinicians often lack predictive tools to guide patient-specific treatment choices.
Meanwhile, patients are increasingly involved in shared decision-making. Bridging the gap between data, practice, and patient needs requires tools that are clinically validated, predictive, and designed for everyday use.

The PHREND® Approach: Predictive Models, Clinical Usability, Regulatory Readiness. PHREND® is a certified Clinical Decision Support System (CDSS), developed by Rewoso in partnership with NeuroTransData. It brings together advanced modeling, clinical relevance, and regulatory robustness to deliver personalized therapy recommendations based on real-world data.

PHREND® combines:

Data Science and Predictive Modeling
Built on a real-world dataset of over 25,000 MS patients treated by more than 100 neurologists, PHREND® applies machine learning and causal inference to model individualized treatment outcomes.
Clinical Integration
Physicians receive ranked therapy options tailored to patient characteristics, medical history, and stated preferences. The interface is designed for daily use – simple, fast, and usable in patient consultation settings.
Regulatory Implementation and Maintenance
PHREND® is CE-certified, GDPR-compliant, and actively maintained for clinical use across Europe. This ensures reliability and legal usability as a certified medical product – not just an algorithm.

Summary of Evidence

PHREND® is backed by independent research and peer-reviewed publications confirming its impact in real-world practice:

Publication 1: Technical Framework Framework for personalized prediction of treatment response in relapsing remitting multiple sclerosis

Published in 2020, this foundational study introduced a Bayesian-based predictive framework using real-world data from the NTD MS registry to estimate individualized treatment response in RRMS. It focused on predicting relapses and confirmed disability progression (CDP) across six DMTs and demonstrated robust performance and generalizability.

Publication 2: Clinical Application
PHREND®- a real-world data-driven tool supporting clinical decisions to optimize treatment in relapsing-remitting multiple sclerosis

Braune, S., | Stuehler, E., | Heer, Y., | van Hoevell, P.,| Bergmann, A., & NeuroTransData Study Group. (2022) Frontiers in Digital Health

This publication presents the technical model into a practical web-based decision-support tool (PHREND®), integrating real-time predictions into clinical routines. It validated the model’s clinical relevance by showing better outcomes in patients who followed PHREND®-recommended therapies.

Publication 3: External Validation Framework for personalized prediction of treatment response in relapsing-remitting multiple sclerosis: a replication study in independent data

Sakr, A. M., | Mansmann, U., | Havla, J., | Ön, B. I., & Ön, B. I. (2024)

An independent research team replicated and validated the original framework using French OFSEP registry data. The replicated models showed comparable or improved predictive performance, confirming the methodological soundness and transferability of the PHREND® modeling
approach.

Results. This Predictive Precision in Practice
PHREND® delivers measurable clinical value:

Personalized therapy recommendations in real-time
Integration of clinical and patient-reported data
Dynamic preference adaptation to reflect real-life patient needs
Automated checklists supporting treatment switching
Deployed in practice by over 100 neurologists and used by more than 25,000 patients in Germany.

Benefits
of the Rewoso
approach

Precision
Modeling

We specialize in building mathematically rigorous models that predict treatment response at the individual level.

Usable Clinical Interfaces

We translate complex data into intuitive tools that doctors can use in their daily practice — seamlessly integrated into real workflows.

Regulatory-Ready Solutions

All tools are CE-certified, GDPR-compliant, and built to meet EU regulatory standards for clinical use — and stay compliant over time.

Accelerated Development

We help partners move faster by building on validated cohorts and proven frameworks like PHREND®.

Beyond Academic Models

We don’t stop at the algorithm. We deliver end-to-end: from AI and modeling to certified, usable medical software.

Benefits of the Rewoso approach

Precision Modeling