Case Studies

Icon
Healthcare UCI,SJHS
Predict patient critical condition in Real-Time

Benefits

  • Proactively Predict events rather than reactively
  • Real-time Alerts
  • Capture & Transmit Patient Vitals at Much Higher Frequencies
  • Improve Patient Satisfaction
  • Improve Operational Efficiency
  • Improved Response Times
  • Reduce adverse Drug Response Times
  • Scope to create real-time and offline models of interest

Problems

Managing the Volumes of System Sensor Data across their Hospital Chain In a typical hospital setting, nurses do rounds and manually monitor patient vital signs. They may visit each bed every few hours to measure and record vital signs but the patient’s condition may decline between the time of scheduled visits. This means that caregivers often respond to problems reactively, in situations where arriving earlier may have made a huge difference in the patient’s wellbeing.

Solutions

  • Logi-Crunch a multi-tenant, scalable healthcare analytics platform that transforms and enriches these sensor data into a manageable dataset
  • Predicts code-blue pathway, septic pathway, CART rule-based-scores in real-time
  • Platform is built on micro-service principles with a plug and play model. Easily extensible to accommodate other predictive models of interest and rules evaluation in real-time
  • Onboarding a new facility, from inception to production, on an average takes about 2 months
  • One of the first few in the world, to build the scalable streaming analytics capability on Apache Nifi
Icon
Healthcare Meharry
DATA Warehouse modernization

Benefits

  • Fosters data-driven decisions
  • Enables ‘schema-on-read’ strategy
  • Low cost on storage and processing
  • Eliminates vendor licensing cost
  • Scope for advanced analytics powered by NoSQL variants

Problems

Legacy system’s large data are growing exponentially. Customer needed a mechanism to reduce the cost, discover business intelligence and discover new revenue streams

Solutions

Our solution resulted in migrating the compete legacy dataset into the Hadoop ecosystem. Process engineered to migrate the data in full-dumps, as well as incrementally. Validation framework, to validate migrated data, metadata and other workloads. Reload the transformed data back to the traditional EDW for cases for specific reporting and to enable phased migration.

Patient De-duplication

Benefits

  • Improves quality of care
  • 360º view of patient information across facilities
  • Enables cohort analysis
  • Lowers probability of repeat tests and treatment delays
  • Aids in precision medicine

Problems

EMR systems ranges between 5-20 percent of duplicate patients record which increases the operational cost. Rate increases to 40% for those hospitals that have merged with other facilities.

Solutions

Logi-MPI is an EMPI engine, powered by a probabilistic patient record matching algorithm. Engine is configurable and the attributes weights could be throttled based on their sensitivity. Engine was run over four of the Hospitals facilities and resulted in 27% match between the patient records across the facilities. Engine also flags probable matches that would need a stewards’ feedback.

Icon
Mining Hindalco
Inventory process improvements

Benefits

  • Inventory strategy by segment
  • Improved obsolescence risk management
  • Inventory flow and bottleneck visibility
  • Casual analysis to improve throughput rate
  • Aid in demand forecasting

Problems

Sub-optimal process integration, unexpected events such as shortage of raw materials and other inventories, 5 – 10% wastage due to obsolescence of inventory, delays in transportation, supply of low grade ore, loss of materials during transportation, reduction in the throughput rate of the processing plats which results in significant financial losses

Solutions

Elogic has proposed a two-phased solution.

  • In the first phase, we embark on studying their processing points and inventory characteristics like resource utilization trends, excess and obsolescence trends, segmentation analysis, replenishment cycles, pilferage analysis.
  • Collect, enrich available historical data of equipment(s) health

In the second phase, Logi-Crunch, our flagship bigdata analytics streaming pipeline will be leveraged to enable –

  • Iteratively, refine and predict failures of critical devices in the supply-chain processes
  • Iteratively integrate and automate current manual processes across the supply chain lifecycle
  • Real-time threshold alerts around replenishments and potential obsolescence
  • Causal analysis of production KPIs