Data-Driven AI transformation in Healthcare
Among others, Covariance Team enables healthcare organizations:
- To unlock the power of health real-world-data to fuel innovation in health and improve patient care is at the heart of healthcare revolution. This is made possible by advances in health technologies and achieved through the extraction of real-world evidence hidden in patients’ medical records.
- To foster research, maximize their operational efficiency and minimize costs, by analyzing large, complex datasets with advanced analytics and machine learning, and the integration of real-time analysis into existing systems and processes.
- To improve customer satisfaction, operational efficacy, financial and administrative performance, increasing revenue and ROI, optimizing resources, and advancing medical research capabilities such as personalized therapies, while in parallel introduces clients to the market of real-world-data, a multibillion market.
- In pharma to identify unmet therapeutic needs, improving RCTs design, accelerating time to market, refining commercial strategies, achieving label expansion, and building clinical-decisions-support systems, while on parallel guide them to their quest in identifying real-world-data sources.
Use cases
Real World Data
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
Increased regulation and competition restricts the success rate of Technology Assessment, and reduces Market Access prices.
Opportunities
Real world evidence significantly increases the success rates for drug market access and costs less that clinical trial evidence. Covariance can provide valuable insights from real world heterogenous data to strengthen a new drug case.
Large-Scale HDCN (Human disease complex networks)
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
The vast majority of DNA tests link a certain disease to one or two specific genes. On the contrary complex diseases are known to be associated with a variety of genes, of which some are linked to other disease respectively. It is therefore sought to map all known associations and their correlation in order to achieve more accuracy in genomic analysis.
Opportunities
AI based analysis of disease-networks in a large-scale perspective
- Extraction of phenotypical knowledge from multiple sources using text mining/NLP.
-Identification and use of well-known biological databases.
-Creation of large disease-networks. Mapping and analysis of common features for drug repurposing.
Predicting Hospital Readmissions (Preventing Readmissions with Data Analytics and AI)
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
Patients with serious and chronic diseases are hospitalized for a period and then being discharged. Unfortunately, multiple studies have shown that up to 25% of these patients will be readmitted within 30 days to be treated again, in some cases with less favorable outcomes. Focusing on value-based care, healthcare providers try to prevent unnecessary readmissions and improve patient care outcomes. Readmission can be significantly reduced by taking steps while the patient is still in the hospital, defining different actions during discharge, and taking steps post discharge to ensure compliance with home care regimens.
Opportunities
AI is ideally adapted in tasks where data inputs are rather complicated and may elude clinicians. In projecting Readmissions risk data are required about the specific patient’s recent care, their current condition, treatment, their home life and other risk factors from patients’ electronic medical records. AI models use this information to provide preemptive assessment of their risk and notify clinicians while the patient is still hospitalized. AI provides reasons that might lead to readmission and provide recommendations for the most likely types of successful treatments according to a patient’s history. The reason codes are valuable to clinicians because they can pinpoint precise areas to focus on when developing a care plan for the patient, thus preventing unnecessary and costly tests.
Medical Testing (Predicting Medical Test Results with Data Analytics and AI)
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
According to the National Academy of Medicine, waste in healthcare is widespread and is estimated to be about $765 billion annually. One of the key areas of waste is unneeded testing or routine tests that are rarely necessary. As stated in ResearchAndMarkets, the global laboratory services market is expected to grow at a compound annual growth rate of 4.21% over the forecast period (2020 -2025), to reach a market size of US$468.249 billion in 2025 from US$365.690 billion in 2019. The number of unnecessary tests result in significant costs that do not necessarily contribute essentially to healthcare quality or positive patient outcomes.
Opportunities
Using Advanced Data Analytics and AI we help clinicians make better decisions by defining the types of tests that are likely to be useful for a patient. One of the advantages of AI is its unique ability to integrate large volumes of data and identify patterns that may be subtle or difficult for humans to recognize. These subtle patterns have an huge potential to alert clinicians to important physiologic changes that need to be addressed. With an AI-driven application we can provide indications of which tests are likely to produce definitive or valuable results based on the patient’s medical history and current symptoms. With this knowledge, the clinician can advise optimal treatments with the best outcomes and minimize the number of tests, which saves time and reduces costs to the patient.
Predicting ICU Transfers (Saving Lives by Catching Patients Before the Crash with Data Analytics and AI)
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
Studies show that patients who undergo an unplanned transfer to the ICU experience worse outcomes than patients admitted directly. These patients typically stay in the hospital 8 to 12 days longer and have significantly higher mortality rates – these patients account for only 5% of patients but represent one-fifth of all hospital deaths. The challenge is to find patients before they “crash” and need to be moved to the ICU, but these patients often lack recognizable symptoms that clinicians can access as ones leading to a serious change in condition.
Opportunities
AI models can be used to find patients who are likely to crash. The machine learning models use patient medical records, laboratory results, and vital signs from patients to find early warning signals of deteriorating condition. These models can be used with existing patients in real-time to determine their risk of a crash and as part of an early warning system for clinicians so they can intervene before the ICU transfer is needed. The AI system can also provide reason codes for a specific patient, which can be a helpful tool to clinicians to understand where they should begin their treatment.
Physician Profiling (Understand physician profiles and clinical performance)
Goal
Strategic Advantage –> Physician Assessment –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
Cost reduction
Quality of services improvement
Optimize staff performance
Opportunities
The economic evaluation & qualitative assessment of a physician’s profile is a measure of quality of Healthcare Service. By using indicators (KPIs) measuring the Physician’s performance in a Healthcare unit we can reduce healthcare costs, improve quality and increase patient satisfaction simultaneously . The AI system can develop a system for KPIs evaluation and desing the necessary intervention steps.
Patient Profiling (Optimization of healthcare services with personalized care management interventions)
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
Patient Evaluation
Cost reduction
Quality of services improvement
Opportunities
Analyze available data, such as:
-Medical services consumer profile
-Geolocation
-Connection with physicians
-Satisfaction from the provided services
-Needs & Preferences
-Optimization of catchment area
-Potentially suboptimal spatial health accessibility of patients.
-Maximization of patient value
-Identify the economic impact of patient navigation in the Healthcare unit to increase revenue per patient through efficient cross, deep & up selling activities.
-Patient Understanding
-Development of innovative products & services based on patient understanding. Using VOC surveys as a tool to reveal customer’s feedback and improve decision making that contribute to a positive return on investment.
Optimization of resources (Optimization of healthcare services with personalized care management interventions)
Goal
Strategic Advantage –> Better/Personalized Treatments –> Customer Satisfaction –> Increased Revenue and ROI
Challenges
Improve utilization
Cost reduction
Operational efficiency & effectiveness of human & asset capital
Optimize supply chain
Optimize the network between the healthcare units & supplies
Opportunities
Analyze all available data such as:
-Staff utilization data
-Asset utilization data
-Suppliers data
-Products usage characteristics
-Products quality
-Products costs
