Our solutions use advanced technology that extends the reach of your knowledge workers without encumbering them with the complexities of what lies underneath and at a price point that is obtainable. Our solutions open the doors for our customers to new ways of understanding, exploiting, and acting upon their data
Demographic and Psychographic classification
Data can be augmented with any type of demographic or psychographic information that can aid in the classification and predictive modeling processes.
Predictive models of future behavior
RelMiner utilizes state-of-the-art predictive modeling based on machine learning technologies. This allows an additional “voice” that can provide proactive and early warning of the possibility of future monitored events. Additionally, this technology is adaptive because it learns over time.
Real time novelty detection and alerting
Unusual or unrecognized patterns can be detected and reported.
Investigation management and prioritization
All investigations are prioritized by an overall score. The score is composed of items that are unusual, of interest to the investigator, or representative of known suspicious patterns. These scores can then be sorted and worked in decreasing order by the investigator based on score thresholds. New patterns can be added. Additionally, certain patterns can be flagged immediately and the entire investigation is boosted in priority.
Maintains a comprehensive audit trail of all activity
All activities are logged so that a complete audit trail can be reviewed of all analytics steps taken, alerts posted, data added, etc. Additionally, commonly used processes and procedures can be “recorded” and replayed against other investigations. This allows for the creation of a catalog of consistent analytical steps that can be reused by team members, and can also be used for training.
PharmaMiner is a system designed to help your organization detect drug diversion, illegal pharmaceutical distribution, Medicaid, and Medicare fraud. Its intuitive design can be easily understood by public and private sector investigators, law enforcement, and attorneys. The service is provided through a secure web interface, making implementation quick and simple. What PharmaMiner Can Do for Drug, Medicaid and Medicare Investigations PharmaMiner can streamline the investigation process by:
- Identifying people, addresses, and other correlated points of interest that exist in multiple investigations. This allows evidence to be combined and results in improved efficiency in investigation and litigation.
- Prioritizing investigations based on the extent of suspicious activity, the amount of money involved in pharmaceutical distribution, and many other criteria specified by the end user.
- Compiling more comprehensive data by incorporating pharmaceutical, provider, pharmacy information, as well as other relevant data sources with standard prescription data as well as other relevant data sources.
- Automatically identifying an extensive set of red flags that take the user deeper into the investigative process thereby making the investigator more productive and accurate.
- Providing extensive reporting that can be easily exported to a variety of case management systems, producing complete file investigation documentation.
CyberMiner is designed to analyze the immense amount of data captured by security logs, Web access logs, email chains, intrusion detection logs, HR records, and many other data sources.
Using CyberMiner, it becomes possible to develop sophisticated profiles of user behavior that can be used to develop predictive models that anticipate future behavior. Work patterns, communication patterns, and outlier analysis all contribute to this model.
CyberMiner uses the same machine learning, graph based analysis algorithms, and Big Data technology as all Anaytics Inside products to provide the following functionality:
- Demographic and Psychographic classification
- Predictive models of future behavior
- Real time novelty detection and alerting
- Investigation management and prioritization
- Maintains a comprehensive audit trail of all activity
Analytic Investigator is a desktop application designed to help your organization manage field investigations. It is easy to use. Our system is easily understood by investigators, law enforcement, and attorneys performing investigative tasks.
Users can use either the Mobile SmartPhone or Tablet based versions, or the desktop version for quick access to all investigation data, and for inputting observations, attaching photographs, and including automatically geocoded location data. It automatically prioritizes investigations. Those with a high number of suspicious “red flags” are emphasized by considering suspicious activities and relationships. Analytic Investigator automatically finds people, addresses, phone number, and other identifiers that show up in multiple investigations. Investigators are notified when these investigations can be combined leading to greatly improved efficiency in investigation and litigation.
Analytic Investigator learns! As investigations are processed, our Knowledge Advisor recognizes and learns from patterns identified in the data. It creates a complete, packaged set of all investigation notes and data. This can be easily exported to a variety of case management systems. This produces an audit trail that is required for successful litigation.
MedMiner™ is an integrated software system designed to help your organization or patient care team:
- Predict patient case outcomes before they occur
- Show patient case trends and alerts that will help direct remediation
- Better comply with ARRA Value Based Purchasing guidelines
- Provide management dashboards that can be used to set guidelines and to track performance
- Extract medical information contained within text documents
MedMiner is a set of products designed to address a broad range of needs in the health care industry. MedMiner is comprised of:
MedPredict™ – Classification and prediction of care outcome.
- It provides dashboard level reporting, early identification of potential undesirable outcomes, and actionable suggestions for remediation. It combines a variety of available data sources including:
- Patient biometric data and historic data
- ICD-9 and ICD-10 diagnosis and procedures
- Pharmaceutical usage
- Patient and facility demographic data
- Patient psychographic data from chart data
- Re-admissions and mortality data
- Inpatient and emergency department discharge data
- MedPredict provides valuable insight to administrators and care providers. Its dashboard allows at-a-glance review of trends and current status. It provides standardization and auditability for both health care plan auditors and for the care team itself.
MedExtract™ – Provides the ability to extract medical terminology from text documents. This wealth of information is unavailable because cannot be queried
- Clinical information extracted from unstructured text
- Most patient information is stored as unstructured text (>60%)
- E.G., care plans, prescripts, doctor’s observations, patients’ communications, etc.
- From the free form text, MedExtract harvests:
- Diseases, diagnoses, and procedures
- Names, addresses, phone numbers, locations
- Drugs, dosages, and usage
- Patient observations of sentiment (depression, anger, etc.)
- Relationships between any of the above
Using quality metrics, the performance of health care regimens must be measured and benchmarked. Our services help design, organize, and correlate these metrics.
How MedMiner is different ?
- Performs analysis that is not feasible within normal relational databases.
- Uses configurable rules and analytical techniques that allows users to find obscured, complex relationships across multiple records in the database.
- Provides near-real time alerts of the problem areas, thus allowing for preemptive mitigation.
- Can utilize all data available at the same time.
- It is consistently keeping track of all events and updates with pattern matching. It re-analyzes every time new records arrive.
The Health Care Effectiveness Data and Information Set (HEDIS) has become the de facto standard for organizational and program reporting. These metrics evolve annually. We can help you analyze those metrics that are important to your organization:
- Effectiveness, access and availability, and satisfaction with the experience of care
- Plan stability and cost of care
- Use of services and wellness programs
Analysis in the health care arena is complex. We can help you model this information in a flexible manner that provides deep insight and adaptability. We specialize in:
- Facility, region, population, and provider
- Claims management
- Fraud and abuse
RelMiner Data Generator (RelMinerDG)
RelMiner Data Generator (RelMinerDG) provides an easy to use, highly customizable ability to generate complex data for testing.
RelMinerDG produces one or more comma delimited files containing the generated data. It provides support for simple data elements (age, weight, etc.), specialized data elements (male/female name, address, phone number, race, ethnicity, eye color, etc.), and more complex data elements (medical provider, prescription, patient, URL/domain, purchased item, credit card number, etc.)
Customized named lists of data can also be created and used as a source for data generation. Generated data can either be totally random, or can be controlled by supplying a probability distribution (75% red, 25% blue).
Generation is also controllable by relationship rules between data elements (if owed_amount > 0 then late_fee = 0.1 * full_amount). Sets of records can be generated and controlled through “populations” (i.e. 10% neonatal, 15% juvenile, 55% adult, 20% geriatric).
An intuitive wizard-like interface is also provided to guide the data generation process. Additional utilities are also available for producing scripts that can be used to load a variety of SQL and NoSQL databases.
Please contact us for more information and a demo.