Search

Autonomous Vehicle technology is quickly expanding its market and has found in Silicon Valley, California, a strong foothold for preliminary testing on public roads. In an effort to promote safety and transparency to consumers, the California Department of Motor Vehicles has mandated that reports of accidents involving autonomous vehicles be drafted and made available to the public. The present work shows an in-depth analysis of the accident reports filed by different manufacturers that are testing autonomous vehicles in California (testing data from September 2014 to March 2017). The data provides important information on autonomous vehicles accidents’ dynamics, related to the most frequent types of collisions and impacts, accident frequencies, and other contributing factors. The study also explores important implications related to future testing and validation of semi-autonomous vehicles, tracing the investigation back to current literature as well as to the current regulatory panorama. [ABSTRACT FROM AUTHOR]

Machine learning algorithms, including recent advances in deep learning, are promising for tools for detection and classification of broadband high frequency signals in passive acoustic recordings. However, these methods are generally data-hungry and progress has been limited by challenges related to the lack of labeled datasets adequate for training and testing. Large quantities of known and as yet unidentified broadband signal types mingle in marine recordings, with variability introduced by acoustic propagation, source depths and orientations, and interacting signals. Manual classification of these datasets is unmanageable without an in-depth knowledge of the acoustic context of each recording location. A signal classification pipeline is presented which combines unsupervised and supervised learning phases with opportunities for expert oversight to label signals of interest. The method is illustrated with a case study using unsupervised clustering to identify five toothed whale echolocation click types and two anthropogenic signal categories. These categories are used to train a deep network to classify detected signals in either averaged time bins or as individual detections, in two independent datasets. Bin-level classification achieved higher overall precision (>99%) than click-level classification. However, click-level classification had the advantage of providing a label for every signal, and achieved higher overall recall, with overall precision from 92 to 94%. The results suggest that unsupervised learning is a viable solution for efficiently generating the large, representative training sets needed for applications of deep learning in passive acoustics., Competing Interests: The author has declared that no competing interests exist.

Background: A significant increase in false positive ST-elevation myocardial infarction (STEMI) electrocardiogram interpretations was noted after replacement of all of the City of San Diego's 110 monitor-defibrillator units with a new brand. These concerns were brought to the manufacturer and a revised interpretive algorithm was implemented., Objectives: This study evaluated the effects of a revised interpretation algorithm to identify STEMI when used by San Diego paramedics., Methods: Data were reviewed 6 months before and 6 months after the introduction of a revised interpretation algorithm. True-positive and false-positive interpretations were identified. Factors contributing to an incorrect interpretation were assessed and patient demographics were collected., Results: A total of 372 (234 preimplementation, 138 postimplementation) cases met inclusion criteria. There was a significant reduction in false positive STEMI (150 preimplementation, 40 postimplementation; p < 0.001) after implementation. The most common factors resulting in false positive before implementation were right bundle branch block, left bundle branch block, and atrial fibrillation. The new algorithm corrected for these misinterpretations with most postimplementation false positives attributed to benign early repolarization and poor data quality. Subsequent follow-up at 10 months showed maintenance of the observed reduction in false positives., Conclusions: This study shows that introducing a revised 12-lead interpretive algorithm resulted in a significant reduction in the number of false positive STEMI electrocardiogram interpretations in a large urban emergency medical services system. Rigorous testing and standardization of new interpretative software is recommended before introduction into a clinical setting to prevent issues resulting from inappropriate cardiac catheterization laboratory activations., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Background: Low-income, Mexican-American patients with diabetes exhibit high rates of medication nonadherence, poor blood sugar control and serious complications, and often have difficulty communicating their concerns about the medication regimen to physicians. Interventions led by community health workers, non-professional community members who are trained to work with patients to improve engagement and communication during the medical visit, have had mixed success in improving outcomes. The primary objective of this project is to pilot test a prototype software toolkit called "EMPATHy" that a community health worker can administer to help patients identify the most important barriers to adherence that they face and discuss these barriers with their doctor., Methods/design: The EMPATHy toolkit will be piloted in an ongoing intervention (Coached Care) in which community health workers are trained to be "coaches" to meet with patients before the medical visit and help them prepare a list of important questions for the doctor. A total of 190 Mexican-American patients with poorly controlled type 2 diabetes will be recruited from December 2014 through June 2015 and will be randomly assigned to complete either a single Coached Care intervention visit with no software tools or a Coached Care visit incorporating the EMPATHy software toolkit. The primary endpoints are (1) the development of a "contextualized plan of care" (i.e., a plan of care that addresses a barrier to medication adherence in the patient's daily life) with the doctor, determined from an audio recording of the medical visit, and (2) attainment of a concrete behavioral goal set during the intervention session, assessed in a 2-week follow-up phone call to the patient. The statistical analysis will include logistic regression models and is powered to detect a 50% increase in the primary endpoints., Discussion: The study will provide evidence regarding the effectiveness and feasibility of a software tool to help patients communicate with doctors about problems they face with their medications., Trial Registration: ClinicalTrials.gov NCT02324036 Registered 16 December 2014.

Radiological images contain a wealth of information,such as anatomy and pathology, which is often not explicit and computationally accessible. Information schemes are being developed to describe the semantic content of images, but such schemes can be unwieldy to operationalize because there are few tools to enable users to capture structured information easily as part of the routine research workflow. We have created iPad, an open source tool enabling researchers and clinicians to create semantic annotations on radiological images. iPad hides the complexity of the underlying image annotation information model from users, permitting them to describe images and image regions using a graphical interface that maps their descriptions to structured ontologies semi-automatically. Image annotations are saved in a variety of formats,enabling interoperability among medical records systems, image archives in hospitals, and the Semantic Web. Tools such as iPad can help reduce the burden of collecting structured information from images, and it could ultimately enable researchers and physicians to exploit images on a very large scale and glean the biological and physiological significance of image content.

PACS is widely used in hospitals and is considered a mission critical system for around-the-clock daily clinical operation. Scheduled or unscheduled downtime of the main PACS archive storage or server could potentially cripple the entire PACS operation. This is especially the case in a filmless hospital environment. Therefore, in a downtime event, it is most desirable for users to have only a minimal performance impact without interruption of clinical data flow or loss of data and to have available historical PACS studies. This paper summarizes some of the developments in the design and implementation of a reliable PACS that insures maximum uptime for end users while preserving the integrity of the PACS data and making it available during downtime events. It also details strategy for developing proper clinical workflow contingency procedures when a scheduled downtime event to the main archive storage and server occurs. Specifically, the design and implementation of a fault-tolerant (FT) main archive server, the development of a FT back-up archive using an application service provider (ASP) model, and the clinical experiences while upgrading a main archive server and migrating the stored PACS data to new storage media will be discussed., (Copyright 2002 Elsevier Science Ltd.)

Hill Physicians Medical Group (and its medical management firm, PriMed Management) early on recognized the need for a data warehouse. Management demanded that data from many sources be integrated, cleansed, and formatted. As a first step, an operational data store (ODS) was built and populated with data from the main transactional system; encounter data were added. The ODS has served its purpose well and has whetted management's appetite for more information and faster, more reliable access, all in one location. PriMed hired Annams Systems Consulting (Annams) for this effort. A team was formed, made up of consultants from Annams and members of PriMed's information services (IS) team. The "classical" approach is being taken: enhancing the ODS, which is largely normalized in structure, and integrating data from various sources, along with enforcing business rules. The team is designing and implementing data marts and a "star schema" style of data modeling--a useful tool for management to evaluate results before investing further.

Frustrated with electronic health records systems on the market, a California physician took matters into his own hands. The result is an electronic medical records system in use by 9,000 physicians in about a dozen communities.

Background: California policy makers have long been hampered by a lack of credible information for use in making legislative or administrative changes to address serious problems in the workers' compensation system. In 1993, the legislature directed the California Division of Workers' Compensation to develop a Workers' Compensation Information System (WCIS)., Methods: An advisory committee developed key questions and identified data sources regarding injury/illness incidence, costs, promptness of benefit delivery, adequacy of benefits, satisfaction with services, and other outcomes., Results: Key data elements were identified, mostly from existing mandated reporting forms for employers and physicians, and from standardized medical billings. Data collection will be carried out using: 1) rapid electronic data interchange (EDI) for a minimum number of data elements; 2) electronic collection of data on medical services for a sample of claims; and 3) surveys to address adequacy of benefits, satisfaction with services, return to work, and other outcomes. A state-based repository will analyze data and provide de-identified public use data files., Conclusions: The proposed WCIS will provide information to improve the performance and to increase the accountability of all the participants in California's Workers' Compensation system. More importantly, it will provide information which will allow improved quality in the provision of mandated benefits to injured workers.

Clinical practice guidelines (CPGs) have been used to reduce unnecessary and undesirable variations in clinical practice. Many facilities succeed in developing and using these tools, while others conclude that they are not worth the effort. This article describes the results of a systemwide analysis of the processes a healthcare system used to plan, develop, implement, and measure CPGs after several years of experience. Extensive interviews were conducted at facilities in which the CPG program was active to determine what factors had contributed to its success or failure. The result is a series of recommendations that will facilitate CPG implementation in virtually any clinical setting.

Background: An academic medical center in an increasingly competitive market, the University of California-Davis Medical Center in Sacramento started working with a consulting firm in 1995 to reduce overall operational costs and costs for the clinical processes involved in treating patients with specific conditions., Establishing the Teams: Twelve operational efficiency (OE) teams and five clinical teams were commissioned, with a combined total of nearly one-half of the target cost reduction. The second wave of six clinical teams was simultaneously initiated in late spring 1996., The Improvement Method: The quality improvement process for clinical improvement teams included the review and inquiry method, which enables many pilot experiments to be conducted in parallel by work groups and coordinated by the main task team., Results and Case Studies: Within six weeks of launching, the 12 OE teams achieved their goals and identified savings opportunities of more than $27 million. One OE team, medical records, had set a goal of $514,000 in cost reduction for a three-year period and achieved the first-year goal of $190,000. For a clinical team on interventional cardiology, the clinical benchmark data revealed that the cost per case of providing cardiac catheterization was greater than for all three benchmark groups. These patients, including 270 patients per year, showed a possible savings through process improvement of nearly $1.4 million. From January 1996 through March 1997, the rate of occurrence of complications decreased from 5.5% to 3%., Epilogue: Physicians gradually accepted more responsibility and accountability for controlling and reducing costs, while maintaining their traditional role as advocates for improved patient care.

To practice medicine in the near future, health care providers in the USA need an information infrastructure they do not yet have. We offer a contribution from social science research to discussions of current medical records practices and how health care activity systems may be transformed by the advent of electronic health records. The goal of the paper is to set forth a framework that connects over-arching questions concerning medical informatics systems development with the practical, cultural and conceptual issues involved in transitions from handwritten and other free text documentation to structured entry of medical records to build patient profiles. The research is broadly framed by an interest in how reciprocal modifications of the design and use of an electronic health record are negotiated in an iterative prototyping project. It is conducted as part of a complex multi-disciplinary research and development effort to create an electronic health record prototype for use in the integrated health care delivery environment of the Southern California Kaiser Permanente Medical Care Program.

Background: A cross-industry benchmarking study in which St Joseph's Regional Health Systems participated recommended that the admissions process (that is, preverifying insurance, obtaining demographic information) be completed before patients arrive at the hospital. To apply this best practice at St Joseph's, management established three team efforts. IMPROVING PREVERIFICATION. One team was assigned to improve the insurance preverification process. After cross-training staff to carry out preverification and piloting a electronic, on-line preverification system, the team increased St Joseph's preverification rate from 2% to 70% for scheduled patients., Exploring Seamless Registration: After exploring solutions for an on-line system that will be integrated across the St Joseph system, the team recommended expanding the current computer system to include all entities and to integrate demographic and clinical information. In addition, a subteam reduced the number of registration forms patients must fill out from eight to one. REDESIGNING ADMISSIONS/REGISTRATION. As part of its restructuring effort, St Joseph's is redesigning six core processes, including admissions/registration. This redesign team is currently implementing two recommendations from the benchmarking study: express admitting and centralized scheduling., Lessons Learned: Organizations are advised to understand the process in its current state by collecting data and working toward improvement goals--and to be prepared for changes in job functions and allocation of resources. Other recommendations are offered to organizations attempting to apply best practices from a benchmarking study.

The software engineering and quality assurance disciplines are a requisite to the design of safe and effective software-based medical devices. It is in the areas of software methodology and process that the most beneficial application of these disciplines to software development can be made. Software is a product of complex operations and methodologies and is not amenable to the traditional electromechanical quality assurance processes. Software quality must be built in by the developers, with the software verification and validation engineers acting as the independent instruments for ensuring compliance with performance objectives and with development and maintenance standards. The implementation of a software quality assurance program is a complex process involving management support, organizational changes, and new skill sets, but the benefits are profound. Its rewards provide safe, reliable, cost-effective, maintainable, and manageable software, which may significantly speed the regulatory review process and therefore potentially shorten the overall time to market. The use of a trial project can greatly facilitate the learning process associated with the first-time application of a software quality assurance program.

The design, development and implementation of medical education software often occurs without sufficient consideration of the potential benefits that can be realized by making the software network aware. These benefits can be considerable and can greatly enhance the utilization and potential impact of the software. This article details how multiple aspects of the IMMEX problem solving project have benefited from taking maximum advantage of LAN resources.

We have designed and implemented a computerized Intensive Care Unit (ICU) decision support alerting system which analyzes all incoming laboratory and blood gas data for critically abnormal values and trends. A computerized patient data management system (HP 78709A PDMS, Hewlett-Packard Co., Waltham, MA) serving 20 Surgical ICU beds is networked to a Clinical Laboratory Information System and a blood gas computer system. The ALERTS subsystem operates on the PDMS as an automatic program triggered by the receipt of fresh laboratory data. Three types of ALERTS are detected: (1) high and low critical values, (2) calculation-adjusted critical values, and (3) critical trends. Once detected, a specific ALERT message is displayed at the bottom of the patient's bedside PDMS terminal and at the central station. Over an eight month period a total of 1,515 ALERTS were detected from amongst approximately 115,000 laboratory data results transmitted to the Surgical Intensive Care Unit (SICU). Slightly over half of all ALERTS were caused by critical blood gas values. ALERTS were found to be a sensitive indicator of severity of illness: patients with one or more ALERTS suffered an ICU mortality of 9.52%, compared to 0% mortality in patients with no ALERTS. We conclude that automated laboratory data ALERTS represent a valuable decision support tool for the management of high risk ICU patients.

Routinely collected vital statistics mortality data (death certificate data) on occupation and industry are useful for (1) generating hypotheses about potential occupational hazards and (2) identifying occupational mortality differentials possibly associated with socioeconomic and life-style factors. This paper presents a Fortran program that analyzes occupational mortality using vital statistics and census data. The user can form any desired grouping of age, race, occupation (or industry), and cause of death codes for analysis. The program also allows stratification on social class or other user-defined correlates of occupation such as smoking behavior. Furthermore, program output can be used for Poisson regression analysis of mortality rates.