RESUMEN

The "Taxonomy of Artificial Intelligence for Medical Services and Procedures" became part of the Current Procedural Terminology (CPT®) code set effective January 1, 2022. It provides a framework for discrete and differentiable CPT codes which; are consistent with the features of the devices' output, characterize interaction between the device and the physician or other qualified health care professional, and foster appropriate payment. Descriptors include "Assistive", "Augmentative", and "Autonomous". As software increasingly augments the provision of medical services the taxonomy will foster consistent language in coding enabling patient, provider, and payer access to the benefits of innovation.

RESUMEN

Drawing on a landscape analysis of existing data-sharing initiatives, in-depth interviews with expert stakeholders, and public deliberations with community advisory panels across the U.S., we describe features of the evolving medical information commons (MIC). We identify participant-centricity and trustworthiness as the most important features of an MIC and discuss the implications for those seeking to create a sustainable, useful, and widely available collection of linked resources for research and other purposes.

Asunto(s)

RESUMEN

The transition from a system focused on the delivery of sickness and illness services to one with a heavy focus of helping people become and remain healthier requires a major shift in how we view the patient and person. The health care system attempts to magically transform us from persons to patients in a context of sickness and disease, in need of medical procedures and interventions. Those few hours we spend a year in formal medical and health care contexts do not define us in the broader life space. We contend that "person-centricity" is more reflective of the life space and as such better supports that shift than do models of consumer or patient empowerment, centeredness, engagement, or activation. "Person-centricity" represents the complexity of how individuals make decisions including health and health care decisions, within the broader context of their lives, and accurately addresses the needs and aspirations of people throughout their life journey. This is not simply a shift in semantics, but an entirely new paradigm that frees the individual from assuming and succumbing to the passive and subservient patient role and dramatically changes the way in which we view ourselves and interact with the health care system.The changes required to create a healthier America and affect costs associated with lifestyle-related diseases need to happen on a personal level, coupled with a supportive infrastructure and public policies to promote and sustain them. This shift is critical to our transition from health care to a healthier way of living and of controlling avoidable costs.

Asunto(s)

RESUMEN

Rapid and affordable tumor molecular profiling has led to an explosion of clinical and genomic data poised to enhance the diagnosis, prognostication and treatment of cancer. A critical point has now been reached at which the analysis and storage of annotated clinical and genomic information in unconnected silos will stall the advancement of precision cancer care. Information systems must be harmonized to overcome the multiple technical and logistical barriers to data sharing. Against this backdrop, the Global Alliance for Genomic Health (GA4GH) was established in 2013 to create a common framework that enables responsible, voluntary and secure sharing of clinical and genomic data. This Perspective from the GA4GH Clinical Working Group Cancer Task Team highlights the data-aggregation challenges faced by the field, suggests potential collaborative solutions and describes how GA4GH can catalyze a harmonized data-sharing culture.

Asunto(s)

RESUMEN

The recent announcement of the Precision Medicine Initiative by President Obama has brought precision medicine (PM) to the forefront for healthcare providers, researchers, regulators, innovators, and funders alike. As technologies continue to evolve and datasets grow in magnitude, a strong computational infrastructure will be essential to realize PM's vision of improved healthcare derived from personal data. In addition, informatics research and innovation affords a tremendous opportunity to drive the science underlying PM. The informatics community must lead the development of technologies and methodologies that will increase the discovery and application of biomedical knowledge through close collaboration between researchers, clinicians, and patients. This perspective highlights seven key areas that are in need of further informatics research and innovation to support the realization of PM.

Asunto(s)

Asunto(s)

RESUMEN

Part-of-speech (POS) tagging is a fundamental step required by various NLP systems. The training of a POS tagger relies on sufficient quality annotations. However, the annotation process is both knowledge-intensive and time-consuming in the clinical domain. A promising solution appears to be for institutions to share their annotation efforts, and yet there is little research on associated issues. We performed experiments to understand how POS tagging performance would be affected by using a pre-trained tagger versus raw training data across different institutions. We manually annotated a set of clinical notes at Kaiser Permanente Southern California (KPSC) and a set from the University of Pittsburg Medical Center (UPMC), and trained/tested POS taggers with intra- and inter-institution settings. The cTAKES POS tagger was also included in the comparison to represent a tagger partially trained from the notes of a third institution, Mayo Clinic at Rochester. Intra-institution 5-fold cross-validation estimated an accuracy of 0.953 and 0.945 on the KPSC and UPMC notes respectively. Trained purely on KPSC notes, the accuracy was 0.897 when tested on UPMC notes. Trained purely on UPMC notes, the accuracy was 0.904 when tested on KPSC notes. Applying the cTAKES tagger pre-trained with Mayo Clinic's notes, the accuracy was 0.881 on KPSC notes and 0.883 on UPMC notes. After adding UPMC annotations to KPSC training data, the average accuracy on tested KPSC notes increased to 0.965. After adding KPSC annotations to UPMC training data, the average accuracy on tested UPMC notes increased to 0.953. The results indicated: first, the performance of pre-trained POS taggers dropped about 5% when applied directly across the institutions; second, mixing annotations from another institution following the same guideline increased tagging accuracy for about 1%. Our findings suggest that institutions can benefit more from sharing raw annotations but less from sharing pre-trained models for the POS tagging task. We believe the study could also provide general insights on cross-institution data sharing for other types of NLP tasks.

Asunto(s)

RESUMEN

This paper describes Kaiser Permanente's (KP) enterprise-wide medical terminology solution, referred to as our Convergent Medical Terminology (CMT). Initially developed to serve the needs of a regional electronic health record, CMT has evolved into a core KP asset, serving as the common terminology across all applications. CMT serves as the definitive source of concept definitions for the organization, provides a consistent structure and access method to all codes used by the organization, and is KP's language of interoperability, with cross-mappings to regional ancillary systems and administrative billing codes. The core of CMT is comprised of SNOMED CT, laboratory LOINC, and First DataBank drug terminology. These are integrated into a single poly-hierarchically structured knowledge base. Cross map sets provide bi-directional translations between CMT and ancillary applications and administrative billing codes. Context sets provide subsets of CMT for use in specific contexts. Our experience with CMT has lead us to conclude that a successful terminology solution requires that: (1) usability considerations are an organizational priority; (2) "interface" terminology is differentiated from "reference" terminology; (3) it be easy for clinicians to find the concepts they need; (4) the immediate value of coded data be apparent to clinician user; (5) there be a well defined approach to terminology extensions. Over the past several years, there has been substantial progress made in the domain coverage and standardization of medical terminology. KP has learned to exploit that terminology in ways that are clinician-acceptable and that provide powerful options for data analysis and reporting.

Asunto(s)

RESUMEN

The deployment of sophisticated software tools and electronic health records offers many new opportunities and challenges to support care delivery. One of the key opportunities is to enhance the quality of care with evidence-based medicine (EBM). One of the key challenges is to embed EBM in tools that directly facilitate the process of documentation and care delivery. Since clinicians typically have the option of using free text for most of their documentation, the tools that provide embedded EBM must be at least as efficient as free text. There are many requirements that must be met in order to effectively embed EBM within clinical content tools and enhance both the usability and the actual use of such tools and clinical content: (1) Facilitate the documentation process; (2) Facilitate the care delivery process, e.g. make order entry faster; (3) Contain recommendations that are highly relevant to the clinical context of an encounter; (4) Aid in the capture of discrete coded data. Support for local variation is often key to meeting these objectives and becomes a central factor in helping clinicians shift from unstructured free text, to the use of these tools, which support the delivery of EBM. This document describes the central tension between the objective of national standardization and delivery of EBM and the need for regional localization of clinical content. This tension must be thoughtfully managed to maximize the quality of care delivery and associated workflow practices. The key elements of legitimate local variation that must be recognized in order to achieve these goals are described in this document, and the key principles for managing the tensions between generalization and localization are identified.

Asunto(s)