Specialized computing environment for co-analysis of proprietary data

Inventors

Hartley, David • Frieder, Ophir

Assignees

Cincinnati Children's Hospital Medical Center • Georgetown University

Abstract

A specialized computing environment that includes hardware and data security features to enable competitive organizations to co-analyze proprietary data without revealing the underlying proprietary data to unauthorized users. Proprietary data are stored in volatile memory, which may be automatically erased according to pre-stored criteria. The analysis is performed automatically by a processing unit without human intervention. Analytical results are sanitized (e.g., using data masking) to prevent the analytical result from being tracible to any particular data source. Sanitized analytical results are output without outputting the underlying proprietary data (except to users authorized to validate analytical results). The computing environment is enclosed within a secure enclosure (e.g., a steel box with a lock), does not include any peripheral devices outside the secure enclosure, does not communicate wirelessly, and does not have hardware ports accessible from outside the secure enclosure (except, in some embodiments, a wired connection for a web server).

Core Innovation

The invention provides a system and method for co-analyzing proprietary data while preventing distribution of the proprietary data. Proprietary data from a plurality of proprietary data sources is stored in non-transitory volatile memory within a secure enclosure that includes a door and a lock, and the computing environment executes instructions stored on non-transitory system memory without human intervention to form an analytical result.

The analytical result is sanitized by changing one or more data elements to form a sanitized analytical result. The sanitization is configured to prevent the sanitized analytical result from being tracible to any data type and any of the proprietary data sources, and only the sanitized analytical result is output or transmitted to one of the proprietary data sources.

The co-analysis of the proprietary data is performed using either an ontology-based approach or a numerical-metrics approach. In the ontology approach, proprietary data is coded according to an ontology, a multi-dimensional ontology space is populated with points corresponding to ontological vectors, populated neighborhoods are identified using an optimization algorithm, and one or more hypotheses corresponding to the populated neighborhoods are identified.

In the numerical-metrics approach, numerical metrics are constructed, a baseline is identified for each numerical metric, additional documents are received and analyzed to identify updated numerical metrics, and updated numerical metrics that deviate from the baseline are identified.

Claims Coverage

The partial content includes three independent claims. Across these independent claims, the main inventive features are directed to a secure-enclosure co-analysis environment with automated execution without human intervention, sanitization that prevents traceability to proprietary data types and sources, and two alternative co-analysis pathways.

Across the independent claims, the coverage is directed to secure-enclosure co-analysis that executes without human intervention, stores proprietary data in non-transitory volatile memory, forms an analytical result, sanitizes the result by changing one or more data elements to prevent traceability to data types and proprietary data sources, and outputs or transmits only the sanitized analytical result to a proprietary data source. The co-analysis is specified as either ontology coding with optimization-based neighborhood hypothesis identification or numerical-metrics construction with baseline and deviation detection.

Stated Advantages

Documented Applications