You are here:  Home | Services | Stratified Medicine Initiatives
Stratified Medicine Initiatives

Cohesion Medical uses stratified medicine approaches to target treatments according to the biological or risk characteristics shared by subgroups of patients.

 

Cohesion Medical software applications can be extended and modified to deliver pathways to Stratified Medicine. By developing predictive learning algorithms it will be possible to perform accurate data-analysis of "patient models" and deliver individualised patient-specific treatments.

 

Outcomes Research Framework (ORF)
All our software have specific Outcomes Research modules built in. It is based on our properietery ORF (Outcomes Research Framework) model which can be integrated with any Clinical Information System capturing patients treatment data. The ORF model adds as an abstract layer over your existing Clinical System and gives you Outcomes Research results without having you to enter any data again. The key benefits of ORF include:

  • No Repeated Data Entry
  • No Learning Cycle
  • Integrates with any Clinical System
  • Easily Exportable Results
  • Extensive Grouping and Querying Options

 

Patient Model
Patient Model or Profiler is a predictive model based on stored patient data that allows comparative analysis of patient details, diseases, treatments and outcomes. On this basis, the model can suggest possible treatments and predicted outcomes for specific patient profiles. The model thus collates and summarises a "clinical prediction" for use by the consultant.

 

Disease Model
Potential for analysing data differently using different mathematical techniques allows predictive algorithms to compile how patterns of psoriasis emerge and evolve over time. The potential for analysing data using different mathematical techniques based on dynamical patterns theory and modern bio-systems theory would allow us to develop an algorithm for analysing how patterns of psoriasis emerge and evolve over time. New mathematical techniques also allow new Measures and Scores to be developed such as a "Disease Criticality Index" to allow screening tests to identify phases of change in patients conditions.

 

Treatment Model
Clinicians can analyse the effectiveness of treatments and medications. Data-analytics can be used to assess the treatment-outcomes variations based on historical data. Using learning algorithms, the system can predict the treatment with most likelihood of successful outcomes for the specific patient.

 

Incidents Model
Incidents can also predicted from "intelligence" gathered from automatic data-analysis of patient records. This can be used as an "early-warning system" by examining large amounts of data for "hidden patterns" indicative of negative outcomes, or severe side effects.