The Guided Diagnostic solution from VIQTOR DAVIS is a Microsoft Azure-based platform that predicts and ranks the likely diagnosis for a human or machine and set of symptoms.
The Guided Diagnostic solution from VIQTOR DAVIS is a proven concept, a blueprint and reusable assets for a Microsoft Azure-based platform that predicts and ranks the likely diagnosis for a given entity ontology and set of symptoms. The scored predictions can be used to inform and augment a human decision maker’s action or trigger a fully automated prescribed action to be taken. The entity being described and presenting symptoms may be a human or a machine, in both cases data that describes each case is likely already captured, the solution leverages this data to predict a likely diagnoses.
How it works Data captured by existing manual diagnostics processes is ingested into the Azure platform. Leveraging the best of Azure technologies, the data is shaped and fed into statistical and Machine Learning algorithms that produce an optimised predictive model which is retrained regularly. The predicted diagnostic data is exposed as output by an API that takes new cases as input. The API is integrated into a decision support system or automated resolution process.
The Use of Accelerators Accelerators feed into every bespoke implementation to accelerate speed to insight. Proven reference architectures for analytical cloud data platforms. Refined templated Azure implementation from which end-to-end environments can be provisioned on demand. Domain-specific knowledge articles that propagate the collective experience and wisdom of global Data Science team.
Project Plan - 50 consultancy days - 10 week plan
5 x days Discovery of Business Problem and Data Analysis
5 x days Design of Technical and Physical Architecture
26 x days Implementation of Azure Environment, Training Data, Define Summary Data, Data Pipeline Development, Data Catalogue, Data Pre-Processing, Build and Train Model
14 x days Evaluate Model Performance, Optimise Model and Test, Deployment Package Model Output, Configure API Endpoint, API Testing, Frontend Integration. Go Live.