Claim Adjudication Analytics

After a claim is submitted, the health insurance company decides how much to pay the healthcare provider. This process is referred to as claims adjudication. The insurer may pay the claim in full, deny the claim, or it can reduce the amount paid to the provider. After an insurer receives a claim, they typically follow a five-step process: initial processing review, automatic review, manual review, payment determination, and payment. 

According to an Accenture study, 75% of insurance executives believe that AI would transform and bring significant changes over the next three years. Automatic adjudication of medical claims can drastically improve how quickly and precisely claims can be processed, leading to higher customer satisfaction. AI makes automatic claims adjudication easier by scanning for errors during the review process and matching key details to simplify the decision of approval or denial.

Neural Network algorithms can help case managers efficiently screen cases, evaluate them with greater precision, and make informed decisions. Hospital claims management is another area that stands to benefit. In a McKinsey study, Germany had an inpatient treatment cost nationwide, which amounted to EUR 73 billion and made up 30 to 40 % of a typical health insurer’s total budget. On average, between 8 and 10 % of all claims received are incorrect. Reliably identifying and correcting these inaccurate claims would save all stakeholders a great deal of time, money, and effort. AI can help achieve this objective. The conventional approach to claims management based on a rule book has been made obsolete by machine learning algorithms that learn from historical cases and continuously evolve. Such a system can systematically identify and correct errors while avoiding unnecessary or ineffective interventions. First estimates indicate that German health insurers could save about EUR 500 million each year.

AI models can play a significant role in highlighting relevant insights about a claim. It can also generate custom alerts based on those insights. With an accurate, transparent, and timely reporting process driven by AI, the customer can feel more confident about the fairness of the claims process and is more likely to accept the settlement offer. The settlement amount would be in line with the right parameters, and hence customers’ interests are safeguarded. Human errors and biases are eliminated.


Germany is a good example of what is happening with health insurance companies. A mid-sized German insurer with over 1.5 million members receives more than 700,000 claims for cost refunds from hospitals every year. Insurers must verify whether the claims are correct – a task that regularly ties down several hundred employees. A McKinsey study shows that almost one in ten claims is incorrect, and the health insurer can challenge the claim’s amount. This process is extremely cumbersome. As a rule, as many as 70 percent of claims are flagged as unusual. This data is based on the health insurer’s specific rule book. Administrative staff then check these claims in detail. Based on the claim information and any available patient history data, the staff then draw on their experience to decide whether to intervene. When genuine claims are rejected or delayed, it can lead to customer dissatisfaction. Claims can also be submitted in physical or electronic format. The main problem with doing things manually is that it requires a lot of manpower. Not only is manual checking time-consuming, it also poses threats like loss of customer base due to delays or perceived harassment over submitted claims. 

The claims adjudication process faces many challenges like delayed reporting, longer cycle times, human error in assessment or filing, fraudulent claims, customer dissatisfaction, and a lack of transparency in the process from the customers’ point of view. It can lead to the loss of several billions of dollars due to fraudulent claims each year that are missed by the inefficient adjudication process.

Solution Approach

Our client wanted to automate the claims adjudication process to improve their efficiency. This was enabled by classifying each item description that was part of the claim into an accurate charge type to apply adjudication rules. platform provides out-of-the-box development frameworks.  The project was started with the relevant environments, which were then created automatically. Development images configured based on pre-defined templates were installed on-premises or in a development VM within the infrastructure. This enabled authentication using LDAP, seamless project setup using Bitbucket, Jenkins, and Docker (ensuring build and deployment without software compatibility issues). 

The platform made available by leverages the latest ML and DL tools while preparing models. It includes Pachyderm-based data versioning, Kubernetes, Kubeflow, and Spark-based ML and DL. It also includes an Istio-based service mesh-enabled microservice architecture, and ELK-based monitoring capability, contributing to a reduction in latency time.

We used NLP and advanced ML-based algorithms to understand the context of each description and prepared a continuous learning system to classify each item into a charge type (e.g., laboratory changes, monitoring charges, etc.) 

Our pre-processing workflow involved using a classification engine to standardize each field by correcting spelling, punctuation mistakes, and expanding abbreviations. After this, we identified the domain similarity of each item with different charge types, which was performed using  advanced ML-based algorithms that were used to classify each item into a one-of-a-kind charge class with a confidence score. 

The details collected were added as exploratory variables by using libraries and analyzed. The attributes obtained were used for categorization (employing Pachyderm-based data versioning) and then performing univariate, bi-variate, and Bag of Words analysis — for both structured and unstructured datasets through xpresso Exploratory Data Analysis (Data and Statistical Analysis).  Different datasets and their different versions were easily controlled and stored into an xpresso Data Model (XDM)-enabled data store that enabled easy retrieval and storage of datasets/ files into internal XDM. This was achieved by using two excellent features of

  1. Data Connectivity Marketplace libraries
  2. Data Versioning

The resulting output was fed back into the application so the application could learn incrementally. The output and each classified item were further re-evaluated with the help of domain experts. Each classified item was passed through a decision node that sent its basis threshold to the business rule engine or for a review from a domain expert. The rule engine that developed incrementally with this classification engine and contract guidelines finally decided claim adjudication. The approach resulted in a 40% increase in assessment speed for claim adjudication and reached over 90% accuracy without dependency on manual review.

How can help Healthcare Organizations transform their journey to cognitive AI solutions is an AI/ML Application Lifecycle Management Platform. enables complete lifecycle management of AI/ML solutions, addressing the AI transformation journey of enterprises on any cloud platform of choice. offers functionality essential for building AI/ML solutions – primarily enabling data scientists to rapidly build predictive and prescriptive models. The platform provides a user-friendly interface to develop, deploy, and manage AI/ML solutions at scale. In addition, supports the incorporation of these solutions into business processes, surrounding infrastructure, products and applications. 

Key benefits of include: 

  • Empowers data scientists to transform AI/ML research into solutions  
  • Improves the productivity of data scientists by enabling them to focus on the business problem, developing algorithms and rapid experimentation of models  
  • Addresses the shortage of skilled data science resources with automated workflows, toolkits and frameworks  
  • Manages AI transformation journey costs without any wastage of R&D efforts  
  • Provides an enterprise-ready and secure environment for complete lifecycle management of AI/ML applications 
  • Enables at-scale deployment of enterprise AI/ML applications on-premise, cloud (AWS, GCP, Azure), or hybrid environments 

Additional details on can be found at: We can schedule a demo of the platform for anyone interested in learning more.

Have Any Questions?

Need more information about the platform?