Healthcare / Anonymous case
Bringing healthcare risk prediction into a clinician-led workflow
Prediction models · Application services · Professional review
Software Sushi delivered prediction models and services, the assessment interface, and backend review workflows for a healthcare project that reached production. The core engineering work was bringing model-generated risk estimates into an application where professionals can review assessments and record their own decisions.
Software Sushi delivered models and prediction services, the assessment interface, and backend review workflows.
The challenge
The challenge: a model needs a workflow
A risk estimate is not the end of a healthcare assessment. It needs to reach the professional reviewing the case, remain connected to the information collected, and stay distinct from the decision that person records.
Our work spanned both sides of that boundary: the prediction capability itself and the application around it. The challenge was to connect modeling, assessment collection, and professional review as parts of the same product.
The core engineering work
Prediction engineering at the center
From models to application services
We built risk-prediction models and exposed prediction functionality through a dedicated API. The services accept questionnaire data and return risk-related estimates for use in the assessment workflow.
That service boundary gives prediction a defined role in the application. The assessment interface collects information; the prediction service produces an estimate; the review workflow lets a professional consider the result and record a decision.
Integrating the estimate, not replacing the reviewer
We connected prediction services with the assessment and review experience. The documented workflow brings model-generated results into professional review alongside the assessment, rather than treating a prediction response as the completed interaction.
The backend supports explicit operations for recording clinical decisions and interventions. This preserves a meaningful distinction in the workflow: what the model estimated and what the professional decided are different pieces of information.
Connected application
The application around the prediction
Structured assessments
We built an assessment interface with conditional questions. A response can reveal a dependent question, while other answers leave that branch hidden. Completion checks apply to visible follow-up questions before the user proceeds.
These interactions connect the structure of the questionnaire to the information passed through the assessment workflow.
Review with context
Our backend work covered assessment records, history, and clinical-review operations. Project documentation also describes configurable follow-up frequencies and aggregate assessment views, extending the application beyond a single submission and result.
The point is a connected review process: collect the assessment, bring the estimate into view, and give professional judgment an explicit place in the application.
Prediction is one responsibility
How the responsibilities connect
- Structured assessment
- Prediction service
- Model-generated risk estimate
- Professional review
- Recorded clinical decision
Prediction models and services
produce risk-related estimates from questionnaire data and expose them to the application.
Assessment interface
collects structured responses and handles conditional questions and completion checks.
Backend review workflows
manage assessment records, history, clinical-decision updates, and follow-up settings.
These are logical responsibilities, not a diagram of a particular cloud deployment or technology stack.
Production delivery
Delivered across the application, taken into production
Software Sushi’s contribution spanned prediction modeling and services, the assessment experience, and backend review workflows. The project reached production.
The engineering result is an application around the prediction—not just an isolated model endpoint. It connects the information being assessed, the estimate being reviewed, and the judgment being recorded.
Responsible clinical boundaries
Risk estimates are inputs for qualified professional review, not autonomous diagnosis or treatment instructions. Production delivery alone does not establish clinical validation, regulatory approval, compliance certification, model accuracy, or improved patient outcomes.
The capabilities described reflect project documentation, not an audited inventory of every deployed version. This case is anonymous and includes no patient records, customer identifiers, or actual application screenshots.
From prediction to product
Bring your prediction capability into a usable workflow
Building a healthcare assessment or decision-support product? We can help connect prediction engineering with the interfaces, application services, and review steps around it.
