Extracting relevant information from physician notes to improve coding and reimbursement. power computer-assisted coding (CAC) and computer-assisted clinical documentation (CDI). Instead of coders carefully reading clinical documentation and producing codes, they can use their expertise to review auto-suggested codes and turn clinical documentation into a rich data repository. Research shows that 80 percent of healthcare data is unstructured. What data are we talking about? Free text found in EHRs, discharge summaries, nursing notes, operative notes, radiology notes and any other dictated or transcribed text left out of a templated report. Appropriate coding relies on proper clinical documentation. That’s why it’s critical to capture complete and accurate clinical documentation up front to prevent disruption in downstream workflow. While coding looks at what the physician said, clinical documentation improvement programs look at what the physician did not say. CDI identifies cases that appear to lack certain elements of documentation. This can be a daunting task. Traditional, manual CDI programs only have the resources to review a sampling of cases, and these limited reviews are often retrospective and don’t always uncover improvement opportunities. A clinically aware NLP extends the reach of CDI programs to all cases and helps pinpoint deficiencies at the point of care. This concurrent review means that issues can be addressed then and there, rather than days or weeks after discharge. More advanced systems prioritize cases by looking at clinical evidence and comparing it to the definitive diagnoses to identify gaps in documentation right away. As a result, the clinical documentation is more likely to accurately reflect the acuity of patient conditions and the course of care during the patient stay

Speed prior-authorization process by identifying relevant medical information. Reducing: laboratory redundancy, imaging redundancy, and administrative costs are a few of the theoretical advantages. Cognitive technologies can help reduce costs by automating tasks, such as reviewing prior authorization requests and de-identifying patient care records, which have historically required human judgment to perform. Prior authorization, a key health care delivery process, requires medical personnel to review treatment requests, clinical guidelines, and health plan policies to decide whether a treatment request should be approved. This largely manual process is costly and time-consuming. Researchers have sought to demonstrate that this process can be automated by applying cognitive technologies. A 2013 study using data from a Brazilian insurer, for instance, applied machine learning—a cognitive analytic technology that uses sets of data to learn patterns and make predictions—to build models of the decision process for approving or denying treatment requests.

Synthesizing medical information into actionable information and care recommendations. interpretation of natural language by CDSS systems will directly assist analysis of primary literature, reducing the need for manually-curated sets of rules that are often limited in topic or scope. By allowing systems such as Watson to analyze information such as patient notes, laboratory data, genotype data, and familial inheritance data, individualized clinical and molecular profiles of each patient can be assembled. The individualized patient profile can then be compared to the thousands of other patient EHRs to identify similarities and associations, thus, elucidating trends in disease course and management. Each new and validated association extends the foundation of the Watson architecture and facilitates increased confidence in subsequent output for diagnoses. These strengths and progress represent a significant advance in the field of clinical informatics.

tag and map unstructured content so that it can interact with structured company data such as fundamentals, capital structures, or sell-side estimates. mine, incorporate, and derive new insights that can supplement or potentially replace traditional factors. explore patterns and trends at a more granular level comparing transcripts and estimates. are companies more likely to call on analysts with favorable ratings after or before a down quarter? Are analysts who participate on company earnings calls more accurate than their peers who do not?

extract information from the unstructured data, such as text within first notice of loss documents, and turn it into structured data. This makes it easy to classify claim submissions for quick resolutions. With the right system of engagement, insurers can automatically check into numerous legacy applications to see if the customers submitting the claim has a policy in good standing and whether they have coverage for this type of loss. For underwriters and brokers, AI can extract and classify information, thus automating the laborious process of data intake, and allowing underwriters to make a go-no go analysis more quickly. Add machine learning to the mix and claims processing technology can learn and guide future decisions.