Suggestions for a web based universal exchange and inference language for medicine. Continuity of patient care with PCAST disaggregation

Implementation of a web based universal exchange and inference language for medicine: Sparse data, probabilities and inference in data mining of clinical data repositories

“…Until recently they have represented a research toolkit but they have recently been integrated into a BioIngine environment being developed to interact with standard medical record systems [2]. DiracMiner faces the biggest challenge, a combinatorial explosion and consequent computational processing and memory burden familiar to data miners, irrespective of the number of records examined (unless an extremely small number).…”

“…with ellipsis '…' means all combinatorial possibilities, P(A), P(B), P(A, C), P(B, D, H) etc. The attributes are separated from each by operators, usually logical [1][2][3][4] and the default, if not shown, is and. As is common, the comma ',' is used for and in theoretical discussion, though in tags it means that the order may be important.…”

“…We previously proposed Q-UEL, a web-based universal exchange and inference language for healthcare and biomedicine [1,2]. Here is it extended to the more traditional domain of public health analysis including general population health sampling, healthcare quality surveys, and screenings.…”

“…Q-UEL's capabilities are already quite broad. Unhampered by any obligations due to existing clinical implementations (but see Section 1.2) and other legacy matters, and free of rigorous and laborious systems of voting for change that characterize the major medical standards efforts [2], we have been able to go back to a "green field", exploring fundamental unifying principles and to build Q-UEL extremely rapidly. Q-UEL communication artifacts are XML-like tags that mostly represent probabilistic statements as captured clinical data and medical knowledge, derived by mining both structured data records and natural language text [1].…”

“…Q-UEL communication artifacts are XML-like tags that mostly represent probabilistic statements as captured clinical data and medical knowledge, derived by mining both structured data records and natural language text [1]. Different Q-UEL tags are concerned with both content and management of data representation including electronic health records and their interoperability and inter-conversion, security and disaggregation [2], knowledge extraction and representation, machine learning [1,3], and hypothesis generation, automated reasoning and decision support [3,4]. The Q-UEL system might reasonably be described as a multivariate architecture for cognitive computing in medicine.…”