Monthly Archives: March 2012

IBM Smarter Analytics Signature Solutions – Predictive Analytics for Healthcare and Other Industries

Posted on by
Reply

By Jane Clabby

We’ve blogged about how healthcare analytics can help improve patient diagnostics and treatment, but another area where analytics can make a huge impact in the healthcare industry in is anti-fraud. Each year healthcare fraud exceeds $250B, according to the FBI.

On March 20, IBM announced new consulting services and software that uses predictive analytics and “big data” to support decision making for the C-suite in three vital areas: Anti-fraud, Waste & Abuse; Next Best Action; and CFO Performance Insight. These solutions are grouped under the family of IBM Smarter Analytics Signature Solutions, and are based on experiences drawn from over 20,000 analytics engagements.

Anti-Fraud, Waste and Abuse uses IBM’s adaptive systems to learn from the latest data, looking for patterns to help protect against emerging fraud.  The solution’s advanced algorithms are able to detect fraud in real-time and recommend the most effective remedy for each case. In some cases, information about an individual might suggest that by just sending a letter requesting payment the issue could be resolved, while in other cases, a more in-depth investigation may be needed.  This solution can be used in healthcare as well as other insurance areas and also to detect tax fraud.

Next Best Action Smarter Analytics Signature Solution is designed to help organizations get a comprehensive view of a particular customer by analyzing traditional customer data as well as customer sentiment data gathered from social media websites such as Twitter. In this way, customer behavior can be predicted as well as customer preferences. This insight can be used by customer service representatives for “upsell” opportunities, by predicting the “next best action” that is most likely to prompt a consumer purchase.

CFO Performance Insight Smarter Analytics Signature Solution helps collect and correlate a range of financial data—this data can provide a much more comprehensive view of business operations across product lines, helping companies determine which products are the most profitable over time, and how to handle inventory levels and distribution to maximize profit.

Natural Language Processing for Comptuer Assisted Coding

Posted on by
Reply

by Susan Eustis

MED2020 supports natural language processing for computer assisted coding in Canada, garnering 50% market share in  health information management solutions in Canada It has a strong ICD10 capability that gives it strategic advantage in the market.  3M is the other Canadian national language processing market participant in the market.  NLP automation of coding by Med2020 supports coder ability to process 100s of charts per day.   Automated systems make coder turn around time faster and processing more efficient.

The  human has to make judgement calls on coding clinician services.  Med2020 prompts someone using natural language processing and providing capability for input using the keyboard or speech recognition.  Speech translates to text and text prompts coders to recognize the text and the ICD10 code.

A coder sitting there can put information into the system directly.  However the coding is done, it recognizes text that describes a service  –  kidney procedure, carpel tunnel diagnosis  –.  The system prompts with a code if it perceives an item that can be coded.  If the code suggested by the computer is correct, that correct ICD10 code is validated by a human and populated back to an abstract where it needs to go to complete the process, automating the coding of clinician services.

The computer assisted coding is saving time and money for a facility using the natural language system.   Operational efficiency is achieved.  This system is close to IBM Watson.   Watson does not yet scan cursive writing, but should have that capability as the technology evolves.   text to using voice recognition for natural language processing.

Automating coding or medical processes is done by prompting the coder with codes and in this manner the computer checks for accuracy.   MED2020 Health Care Software  is heading for really high percentiles for accuracy.      It is being featured on an ehealth innovation showcase in western cites of Canada.

MED2020 Health Care Software Inc., a Canadian pioneer in Health Information Management solutions launched MED2020 codeSmart, a computer-assisted coding tool developed in collaboration with the National Research Council Institute for Information Technology (NRC-IIT).

Using Natural Language Processing (NLP) the system is able to analyze and convert speech to text.  MED2020 codeSmart is designed to facilitate automation of the coding processes for Health Records departments in hospitals across Canada.  Leveraging the considerable expertise of NRC’s research team, MED2020 is positioned to solve the challenges hospitals face in overcoming coding backlogs, coder shortages.  The aim is to improve efficiencies.

The health care industry (with the World Health Organization) is determining the next coding classifications and is looking at ICD 11.  MED2020 in Canada is looking at that initiative and thinking about making preparation for the next wave of specifications.

Memorial Sloan-Kettering Cancer Center and IBM Watson Collaborate in Cancer Treatment

Posted on by
Reply

By Jane Clabby

IBM and Memorial Sloan-Kettering Cancer Center (MSKCC) announced an agreement to collaborate on developing a decision support tool based on IBM Watson technology that will create individualized diagnosis and treatment recommendations for cancer patients. The tool will combine Watson’s computational power and natural language processing with MSKCC’s clinical knowledge, molecular and genomic data, and database of cancer case histories to support evidence-based medicine.

With cancer-related medical information doubling every five years, and hundreds of cancer sub-types being identified, it is a huge challenge for oncologists to stay current with each cancer type, particularly for physicians in remote areas. This complexity of treatment and diagnosis options lends itself to Watson technology which can process and understand 200 million digital pages and deliver an answer within 3-5 seconds. By leveraging Watson’s ability to ingest, process and analyze volumes of unstructured data, cancer therapy can be personalized —speeding treatment and improving care.

Dr. Larry Norton, Deputy Physician-in-Chief for Breast Cancer Programs Medical Director, describes this evidence-based approach enabled by Watson as going “past intelligence into wisdom.”  He believes Watson can function as a “wise counselor” to physicians and that availability of Watson technology represents a revolutionary step that will “change the way we conduct medicine” by moving from opinion-based medicine to evidence-based medicine.

Pilot solutions for lung, breast and prostate cancer will be available to select oncologists in late 2012, with wide distribution expected in late 2013.

An Update on Seton Healthcare Family and Watson

Posted on by
Reply

An Interview with Ryan Leslie, VP of Analytics and Health Economics

By Jane Clabby

On March 20, I attended IBM’s Smarter Analytics Leadership Summit in New York City. I was pleased to discover that the Client Panel, “Leading with Analytics”, included Ryan Leslie, VP of Analytics and Health Economics at Seton Healthcare. I wrote a blog about Seton Healthcare and Watson about a month ago, and this was a great opportunity to hear more about that project.

Seton Healthcare is using Watson technology to improve patient care and reduce patient readmission rates.  According to Leslie, 80% of Seton’s patient data is unstructured, including physician and nurse notes, discharge summaries and case notes. A standard chart review doesn’t reveal any of this information, and in a complex condition such as congestive heart failure which has to be diagnosed early for effective treatment, these details are critical.  By using this unstructured data, Seton Healthcare has been able to institute a program that tells them which patients have the most potential for positive outcomes. By revealing patterns in unstructured patient data, they can identify those patients that will benefit most from a particular treatment plan, and avoid costly readmissions.

The most important information, according to Leslie, is garnered by reading between the lines — looking at patient information such as does the patient have access to transportation?  Do they have a primary care physician? Do they have a social support system? Answers to these questions can identify risk factors for readmission. Leslie said that through the study, Seton Healthcare had identified another risk factor that frequently led to readmission. During physical examinations, patients were asked to tilt their head so physicians could examine the jugular vein. This simple test revealed vein patterns that were frequently seen in readmitted patients. By using this test rather than expensive array of lab tests, Seton Healthcare is better able to identify at-risk patients at a much lower cost.

HP and Lucile Packard Children’s Hospital Pilot Electronic Patient Dashboard

Posted on by
Reply

By Jane Clabby

In May 2011, HP announced a financial commitment of $25 million over ten years to support a research initiative at Stanford’s Lucile Packard Children’s Hospital, ranked one of the nation’s leading  pediatric hospitals by U.S. News and World Report. The goal of the collaboration is to improve the safety and quality of care for critically ill patients. In addition to supporting the expansion of the Packard Children’s Hospital, the investment produced a new real-time patient status system that resulted in more targeted care in 1 out of 3 patients in a trial period, HP reported in October 2011.

The Patent-Centered Dashboard collects data from electronic medical records and displays that data electronically, replacing traditional error-prone handwritten whiteboards. The electronic dashboard collects the wealth of data found in the medical record and uses it to generate patient status information. Color-coded lights indicate the urgency of patient treatment so that staff can act proactively to address life-threatening situations in the intensive care unit. The dashboard also improves compliance and will be rolled out in other hospitals as well.

The dashboard is designed to prevent human error which can increase the risk of patient death or unnecessarily prolong a hospital stay. Doctors and nurses can easily review patient status on a daily basis and update patient treatment. Several examples of where the dashboard has helped improve patient care include switching a patient from intravenous to oral medication, removing a catheter when the potential for infection was identified, identifying overdue procedures, and decreasing the use of unnecessary lab tests. These measures not only improve care but they can also reduce costs.

Collaboration between Physicians and Hospital Administrators is the Key to Better Patient Care

Posted on by
Reply

By Jane Clabby

Dr. Craig Saunders offers his thoughts on improving care delivery

Dr. Craig Saunders, Chairman of the Department of Cardiothoracic Surgery at Barnabas Health System, offered his thoughts on our healthcare system in a recent interview.  He believes a large part of the issue is the disconnect between physicians and hospital administrators. One example he noted is the existence of both a Medicare Part A (geared toward hospitals and medical centers) and a Medicare Part B (geared toward physicians). He went on to say that by working together physicians and hospitals could build something “greater than themselves”, and in the patient’s best interest. With the trend toward physicians being employed by hospitals, the need for closer collaboration is becoming even more critical. The physician, according to Dr. Saunders, should provide the “vision and passion” and the hospital administrators should provide the resources to support that. A successful partnership would be financially responsible on both sides.

Dr. Saunders believes that Watson can play an important role in this closer collaboration and fiscal responsibility.  While physicians cannot possibly stay current with medical analysis and peer reviews conducted on a daily basis, Watson can process and analyze huge volumes of information so that actual evidence-based medicine can be practiced. In this way physicians can eliminate “soft data”, provide more targeted diagnoses and treatment and potentially avoid costly legal issues. Having access to specific patient data and history, correlated with volumes of medical research analyzed by Watson also has the potential to offer more preventative and lifestyle-related medical care, and reduce costly end-of-life care (according to Dr. Saunders as much as 80% of medical costs are incurred in a patient’s last year of life).

Care Delivery In a Services Oriented Facility

Posted on by
Reply

by Susan Eustis

Mayo clinic, Cleveland Clinic, and Lahey Clinic provide a collaborative healthcare services delivery environment that is the preferred model going forward according to Craig R. Saunders, M.D., Chairman, Department of Cardiothoracic Surgery, Barnabas Health System in a private interview with WinterGreen Research this week.  He cited statistics relevant to demonstrating the fact that the US healthcare delivery system is the best in the world, unrivaled by any other care delivery system.  As we work to make the system more inclusive it is important not to destroy what is working well now according to Dr. Saunders.

Those models he cited, Cleveland Clinic, Lahey, and Mayo are where the physicians work collaboratively.  These facilities have been able to produce these best of the best in diagnostic and therapeutic care.   He talked about the necessity of balancing care delivery with resource availability and an adequate flow of income to keep facilities financially sound.

Care decision support represents a way to achieve more efficient and better care delivery.  No one physician or clinician can keep up with all the data coming at them with respect to best practice and medical breakthroughs, even in a particular specialty, so he embraced the idea of natural language based systems that assist and support physicians.  Dr. Saunders is a strong advocate for improving healthcare delivery in a manner that is led by the most thoughtful and intelligent practicing physicians, not leaving healthcare change to politicians.  In this he deserves to be listened to, and we urge him to work with Dr. John Livingstone of the Harvard Medical School who has been a similar advocate nationally for many years.

 

How to Fix our Healthcare System

Posted on by
Reply

By Jane Clabby

A Physician from Barnabas Health System Weighs In

Dr. Craig Saunders, Chairman of the Department of Cardiothoracic Surgery at Barnabas Health System, shares his thoughts on the problems with our health care system.  According to Dr. Saunders, the solution includes leaving politicians out of the equation.

“There is no crisis in the practice of medicine, the crisis is in the business of medicine.  Unfortunately money cannot be taken out of the equation; the payers now run Health Care.  Politicians, corporations, insurance companies, lawyers and accountants now have more influence on health care than patient consumers and physician providers.  Physicians should be chastised for allowing that to happen and for letting their patient advocacy falter.”

“Unfortunately there are probably as many bad physicians as there are bad lawyers but the good ones have to step up and provide the passion and vision for tomorrow’s health care.  It won’t look like your grandfather’s health care . . . and it shouldn’t, the only constant is change. Private practice may be a relic of the “cottage industry” phase of medicine but that doesn’t mean an impersonal government or corporation should run health care either.  The future of medical care should be based heavily on a collaboration of able and motivated physicians and administrators working together to build quality health care programs with fiscal responsibility.  With fiscal responsibility will come sustainability and with the quality will come value.”

“This may all sound theoretical but already excellent examples exist and nothing exists that wasn’t, at one point, theory.”

Stay tuned for future blogs that provide details of our interview with Dr. Saunders.

Natural Language Processing Used for Computer Assisted Coding

Posted on by
Reply

by Susan Eustis

Natural Language Processing Used for Computer Assisted Coding

CodeRyte offers sophisticated technology that automates medical coding in leading healthcare settings including single- and multi-specialty practices, academic medical centers and integrated delivery systems.  A web-based software solution automates the coding process.  A natural language processing engine reads the text of a physician’s report and automatically assigns appropriate codes.   Coded notes are for client review either on-site or remotely through a secure, HIPAA compliant Web interface.

A-Life uses its proprietary and patented Natural Language Processing (NLP) technology, LifeCode, to decipher electronic transcribed patient encounters via the Internet through its data center.  Documents are coded for reimbursement purposes.  A-Life’s NLP technology utilizes proprietary knowledge bases which contain tens of millions of facts to automate the coding process.

A-Life Medical is part of Optum.  OptumInsight specializes in improving the performance of the health system by providing analytics, technology and consulting services that enable better decisions and results.  A-Life Medical computer-assisted coding (CAC) offers advanced computerized coding and coder management solutions.

3M 360 Encompass is an encoding system, a software application that connects and automates coding, clinical documentation improvement (CDI), and performance monitoring in a single integrated workflow.  3M 360 Encompass System helps hospitals and health systems achieve complete documentation, accurate and timely reimbursement, and outcomes data that reflects the care provided.

3M 360 Encompass system addresses the inefficiencies of multiple coding and documentation workflows offering a breakthrough answer to the coding and documentation challenges of healthcare reform and ICD-10.

Cerner Corp. has acquired LingoLogix, a Dallas-based vendor of computer-assisted coding software.  Cerner has changed LingoLogix’s name to Discern nCode and is integrating the coding software with its core Millennium information system.

Discern nCode uses natural language processing technology to identify key words and phrases in physician dictated notes either in a voice file or machine-readable text. The software then matches the key words and phrases to appropriate SNOMED-CT terminology and ICD-9 or CPT-4 codes, and then automatically codes the notes. The notes then are searchable for billing or research purposes.

 

 

Steve Mills Speaking at IBM Pulse: Running the Business is About Sharing

Posted on by
Reply

by Susan Eustis

Steve Mills at IBM Pulse:  Running the Business is About Sharing

Steven A. Mills, IBM Senior Vice President and Group Executive – Software & Systems speaking at Pulse on March 6, 2012, invited the audience of developers to rethink the cultural model for IT.  He spoke about the necessity of sharing in all aspects of a business.   Steve Mills emphasized that sharing is fundamental to process and cost takeout.   He said that formulating IT, running automated process is a journey.   “Rethink the cultural model for IT and you will get value out of it,” said Steve Mills.

Steve Mills was inviting the Pulse developer audience of 8,000 to rethink the culture model for IT to surface the cultural conflicts in IT that exist between distributed systems personnel and mainframe centric personnel.  Mills challenged the audience of developers to understand that sharing is fundamental to teamwork and efficient operations.    Virtualization of infrastructure is the new model for sharing and that technology has touched every IT department.

Information technology is impacted by the economics.  Smarter planet is an IBM initiative built to address managing the resources brought by the challenges of effectively using 1.2 zettabytes of data.   Steve Mills talked about the rate of change in IT and the use of great analytics to address understanding data and turning it into information using dashboards.  The single view of the data was a core theme of Steve Mills speech.

Steve Mills talked about IT In a world where there are 32.6 million servers  –  operating at 85% idle capacity.  The energy power and cooling costs for server farms are improved by implementing shared systems.   Management and administrative costs are controlled when shared computing environments are put in place.   New server spending can be controlled with virtualization, based on leveraging the mainframe in combination with distributed servers to implement a hybrid modernized computer infrastructure.  The idea is not to go completely to the mainframe, but to the extent possible to leverage the positive parts of that systems implementation that provides a flexible hybrid systems that can be provisioned in an automated manner.

IBM uses all different platforms; the applications are put on the platform that is most appropriate for that particular task.  That is the best way to control labor and systems costs.  IBM is investing in helping manage IT costs that are jumping off the charts.  Management of server sprawl drives cost.  By improving the ability to manage  better, it is possible to address problems associated with consuming an ever larger proportion of IT budgets by supporting achieving ways to operate more efficiently

IBM corporate infrastructure is comprised of 110 petabytes of data that are handled with 1.2 million MIPS of capacity.   As IBM sought to deal with these issues internally, servers were consolidated.  IBM achieved a reduction in the number of applications from 15,000 applications to 4,500 applications.  This represented a cultural change because every part of the IBM organization swore that they needed each of the 15,000 applications.  No one wanted to let go of their own application.

IBM has been able to internally increase storage utilization from 50% to 90 % achieving effective TCO.   The value of mainframe is apparent as distributed workloads are moved to a  faster processor on the main frame.    Linux runs on the main frame as well as it does on distributed servers.  Customers can save money on software when they consolidate workloads on the mainframe.

Systems consolidation implementation on a range of platforms is implementing modernization.  IBM has spent a lot of time looking at ways to manage bringing down systems cost while maintaining or improving service levels.  Sharing of information and data addresses the issues of getting the right information to the right place at the right time.  New products represent a way to model one common expression of data and share it across applications that require more data from a variety of sources.  Steve Mills made a compelling case for implementing systems in a manner that is thoughtful and addresses issues of sharing and culture in a manner that delivers value to the business.