SEARCH · PubMed Health

Results for “Clinical workflow”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 649 records · Page 36Linked to original sources

PubMed2002

Evidence-based careflow management systems: the case of post-stroke rehabilitation.

The activities of a care providers' team need to be coordinated within a process properly designed on the basis of available best practice medical knowledge. It requires a rethinking of the management of care processes within health care organizations. The current workflow technology seems to offer the most convenient solution to build such cooperative systems. However, some of its present weaknesses still require an intense research effort to find solutions allowing its exploitation in real medical practice. This paper presents an approach to design and build evidence-based careflow management systems, which can be viewed as components of a knowledge management infrastructure each health care organization should be provided with to increase its performance in delivering high quality care by efficiently exploiting the available knowledge resources. The post-stroke rehabilitation process has been taken as a challenging care problem to assess our methodology for designing and developing careflow management systems. Then a system was co-developed with a team of rehabilitation professionals who will be committed to use it in their daily work. The system's main goal is to deliver a full array of rehabilitation services provided by an interdisciplinary team. They are related to identify which patients are most likely to benefit from rehabilitation, manage a rehabilitation treatment plan, and monitor progress both during rehabilitation and after return to a community residence. A model of the rehabilitation process was derived from an international guideline and adapted to the local organization of work. It involves different organizational units, such as wards, rehabilitation units, clinical laboratories, and imaging services. Several organizational agents work within them and play one or more roles. Each role is defined by the goals' set that she/he must fulfill. Special effort has been given to the design and development of a knowledge-based system for managing exceptions, which may occur in daily medical work as any deviation from the normal flow of activities. It allows either avoiding or recovering automatically from expected exceptions. When they are not expected, organizational agents, with enough power to do that, are allowed to modify the scheduled flow of activities for an individual patient under the only constraint of justifying their decision. After an intensive testing in a research laboratory, the system is now in the process of being transferred in a real working setting with the full support of its future users.

Delivery of Health Care↗

PubMed2002

Real-time computerized in situ guidance system for ACL graft placement.

A recent consensus within an international society for sports traumatology revealed that approximately 40% of ACL grafts are being surgically misplaced in current clinical practice. To help solve this problem, a computer-assisted system has been developed at the M.E. Müller Institute for Biomechanics to perform intraoperative planning and guidance of ACL replacement. Dynamic reference bases are fixed on the femur and tibia to track the knee's movement. No intraoperative imaging is required, and potential ligament attachment sites can be directly digitized using a computerized palpation hook in a minimally invasive fashion when used in conjunction with standard endoscopic tools. The palpation hook can be used by the surgeon to interactively define various anatomical structures and reference landmarks that are important for proper ligament positioning. The system can input a standard diagnostic X-ray (sagittal view of the femur) and allows intraoperative registration of this image with the patient to provide valuable X-ray landmarks for intraoperative guidance. The computer helps in situ planning of ligament placement by providing the surgeon with a 3D overview of the relevant anatomical landmarks and information on graft impingement and elongation for various simulated surgical insertions and graft sizes. After planning, the computer helps guide placement of the chosen insertion tunnels. This approach provides an augmented 3D view of knee anatomy and ligament function prior to drilling that is not possible with current procedures. The flexibility of the system in permitting surgeon-defined landmarks and free interpretation of functional factors allows it to support a variety of surgical workflows and techniques.

Anterior Cruciate Ligament↗

PubMed2002

RIS minus PACS equals film.

Web-based integration methods can be used to resolve a fundamental issue in the transition from film to a picture archiving and communication system (PACS): the identification of relevant prior studies only available on film. Even in the most ambitious conversions from a film-based environment to PACS, there are issues regarding prior studies not on PACS. Failure to compare with prior exams is one of the known risk factors for malpractice in radiology. While most commercial PACS systems today have some degree of RIS integration, knowledge of prior studies is usually limited to an awareness of studies in the PACS. On the other hand, most RIS systems today do not or cannot distinguish between studies on film and those in PACS. We made the observation, from a set theory perspective, that in general: Therefore we sought to create a system that would query both the RIS and PACS and reconcile the results using the above set operation. The query is initiated from a display station via the invocation of a Web browser installed on that station. The process of starting the browser is implemented using a scripting language provided by the workstation vendor, though the use of other mechanisms, such as the CCOW (Clinical Context Object Workgroup) or IHE (RSNA Integrated Healthcare Enterprise) interfaces, can be supported by this architecture. The medical record number, which identifies the current patient and is the primary parameter of the query, is passed as part of the URL (Universal Resource Locator) used to launch the browser. Once running, the browser connects to a Web server that hosts a JSP (Java Server Page) page that performs a DICOM query of the PACS and an HL7 query of the RIS, and then collates the results using the set operation described above. Both the DICOM and HL7 functionality are provided by Java-based toolkits developed in house. The results are returned to the client's browser as a standard HTML page with a tabular format detailing which studies are on PACS and which are available only on film. The responsiveness of the system in terms of time required to complete the two queries and display the results was measured. In addition, the number of diagnostic reports, whose retrieval was triggered by the results of the queries, was monitored to determine the overall performance and use of the system. This project demonstrates that, with minimal modification of commercial software, Web-based integration methods exist to enable patient-context sensitive queries from the diagnostic workstation that identify relevant prior studies that exist only on film and are unknown to the PACS. As a result radiologist workflow is enhanced by the elimination of the need to consult a physically separate system for this type of information. In addition, quality of service is improved by providing more accurate and easier identification of relevant prior studies.

Information Storage and Retrieval↗

PubMed2006

[Quantification of pulmonary emphysema in multislice-CT using different software tools].

PURPOSE: The data records of thin-section MSCT of the lung with approx. 300 images are difficult to use in manual evaluation. A computer-assisted pre-diagnosis can help with reporting. Furthermore, post-processing techniques, for instance, for quantification of emphysema on the basis of three-dimensional anatomical information might be improved and the workflow might be further automated. MATERIALS AND METHODS: The results of 4 programs (Pulmo, Volume, YACTA and PulmoFUNC) for the quantitative analysis of emphysema (lung and emphysema volume, mean lung density and emphysema index) of 30 consecutive thin-section MSCT datasets with different emphysema severity levels were compared. The classification result of the YACTA program for different types of emphysema was also analyzed. RESULTS: Pulmo and Volume have a median operating time of 105 and 59 minutes respectively due to the necessity for extensive manual correction of the lung segmentation. The programs PulmoFUNC and YACTA, which are automated to a large extent, have a median runtime of 26 and 16 minutes, respectively. The evaluation with Pulmo and Volume using 2 different datasets resulted in implausible values. PulmoFUNC crashed with 2 other datasets in a reproducible manner. Only with YACTA could all graphic datasets be evaluated. The lung volume, emphysema volume, emphysema index and mean lung density determined by YACTA and PulmoFUNC are significantly larger than the corresponding values of Volume and Pulmo (differences: Volume: 119 cm(3)/65 cm(3)/1 %/17 HU, Pulmo: 60 cm(3)/96 cm(3)/1 %/37 HU). Classification of the emphysema type was in agreement with that of the radiologist in 26 panlobular cases, in 22 paraseptalen cases and in 15 centrilobular emphysema cases. CONCLUSION: The substantial expenditure of time obstructs the employment of quantitative emphysema analysis in the clinical routine. The results of YACTA and PulmoFUNC are affected by the dedicated exclusion of the tracheobronchial system. These fully automatic tools enable not only fast quantification without manual interaction, but also a reproducible measurement without user dependence.

Adolescent↗

PubMedDate not supplied

Factors affecting and affected by user acceptance of computer-based nursing documentation: results of a two-year study.

OBJECTIVES: The documentation of the nursing process is an important but often neglected part of clinical documentation. Paper-based systems have been introduced to support nursing process documentation. Frequently, however, problems such as low quality of documentation are reported. It is unclear whether computer-based documentation systems can reduce these problems and which factors influence their acceptance by users. DESIGN: We introduced a computer-based nursing documentation system on four wards of the University Hospitals of Heidelberg and systematically evaluated its preconditions and its effects in a pretest-posttest intervention study. For the analysis of user acceptance, we concentrated on subjective data drawn from questionnaires and interviews. MEASUREMENTS: A questionnaire was developed using items from published questionnaires and items that had to be developed for the special purpose of this study. RESULTS: The quantitative results point to two factors influencing the acceptance of a new computer-based documentation system: the previous acceptance of the nursing process and the previous amount of self-confidence when using computers. On one ward, the diverse acceptance scores heavily declined after the introduction of the nursing documentation system. Explorative qualitative analysis on this ward points to further success factors of computer-based nursing documentation systems. CONCLUSION: Our results can be used to assist the planning and introduction of computer-based nursing documentation systems. They demonstrate the importance of computer experience and acceptance of the nursing process on a ward but also point to other factors such as the fit between nursing workflow and the functionality of a nursing documentation system.

Attitude of Health Personnel↗

PubMed1986

Application of network planning methodology to workflow analysis in a reference laboratory.

Efficient organization of the workflow in a reference laboratory is essential to timely resolution of serological problems. This study investigated the usefulness of network planning methods to streamline workflow in a blood center reference laboratory. The investigation of a patient serological problem which included a positive direct antiglobulin test was chosen for analysis. Individual steps in the resolution process were identified. All possible logical sequences of these steps were investigated to determine which steps were along the 'critical path', the sequence of steps which is rate-limiting. By implementing new procedures for these steps, faster resolution of a problem should result. The Program Evaluation Review Technique predicted that altering procedures to shorten 3 steps (autoadsorption, antibody identification panel at 37 degrees C and screening for antigen-negative units) would result in a decrease in the time required to complete a patient problem by 29%. Time-motion studies documented a 35% improvement in problem resolution time with the new procedures (p greater than 0.05). Utilization of the network planning techniques in an immunohematological laboratory may allow for more efficient and expedient operation.

Clinical Laboratory Techniques↗

PubMed2006

The feasibility of implementing an electronic prescribing decision support system: a case study of an Australian public hospital.

Medication errors are common in public hospitals, with the majority at the prescribing stage of the medication pathway. Electronic prescribing decision support (EPDS) is a rules-based computer system that can be used by clinicians to warn against such errors to improve patient safety and support staff workflows. Despite its apparent advantages, this technology has not been widely adopted in Australian public hospitals for inpatient prescribing. A case study using Sauer's (1993) Triangle of Dependencies Model was conducted in 2003 into the feasibility of implementing an EPDS system at an Australian public hospital in New South Wales. It was found not feasible to implement an EPDS at the hospital studied due to the legacy patient administration system, low availability of information technology on the wards, differing stakeholder views, legislation, and the Independent Pricing and Regulatory Tribunal of NSW report recommendations. A statewide standard was preferred, with an agreed specification framework identifying basic core data items and functions that an EPDS must meet which can then be used by area health services to: (i) choose a solution which best meets their contextual needs; and (ii) engage vendors to tender for building an open source (non-proprietary) system based on the specification framework.

Adolescent↗

PubMed2004

Organizing the transfer of patient care information: the development of a computerized resident sign-out system.

BACKGROUND: The problem of safe and efficient transfer of care has increased over the years as new and complex diagnostic tools and more complex treatment options became available. Traditionally, residents ensured continuity of care by working long hours and minimizing the transfer of significant diagnostic or therapeutic responsibilities to other providers. The new 80-hour workweek has curtailed that practice and increased the pressure on trainees for workflow efficiency. We report on a study of information-handling routines among residents for the separate tasks of transfer of care ("sign-out") and daily patient care work (ward work). Using these results, an institution-wide computerized system was developed to centralize information-handling tasks and facilitate the management and transfer of patient care information. STUDY DESIGN: House staff from 31 resident-run inpatient and consult services at 2 teaching hospitals described current methods of maintaining patient information used during ward rounds and during sign-out. A subgroup of 28 residents then participated in the design of a computerized resident sign-out system to centralize patient information and produce lists for rounding and transferring care duties. Accuracy, flexibility, and portability were identified as key elements by the design team. RESULTS: Analysis of the type of information handled by residents caring for inpatients at our institution demonstrated common elements across many services. Most services used a paper patient list to manage both nightly sign-out and daily ward work, which required repeated recopying of patient data during the day. Utilizing medical information systems tools and rapid application development concepts, we constructed a computerized resident sign-out system ("UWCores"). This system combines the patient sign-out and daily ward work information in one central location. We believed this would improve the quality of information transferred during sign-out and enhance resident efficiency. During the design process, we identified rules that govern the type of clinical information that should be automatically versus manually updated. We observed an immediate acceptance by all residents and services that tried the system. CONCLUSIONS: This study shows that by combining downloaded patient data from hospital systems with resident-entered patient details, a computerized resident sign-out system can be a feasible, powerful, and popular tool. While its effect on patient safety and resident efficiency await the results of further studies, our study shows that this tool rapidly captured the attention of resident physicians and became widely used as a valuable means to centralize and organize sign-out and daily ward work information.

Continuity of Patient Care↗

PubMed2025

Building a Digital Health Research Platform to Enable Recruitment, Enrollment, Data Collection, and Follow-Up for a Highly Diverse Longitudinal US Cohort of 1 Million People in the All of Us Research Program: Design and Implementation Study.

BACKGROUND: Longitudinal cohort studies have traditionally relied on clinic-based recruitment models, which limit cohort diversity and the generalizability of research outcomes. Digital research platforms can be used to increase participant access, improve study engagement, streamline data collection, and increase data quality; however, the efficacy and sustainability of digitally enabled studies rely heavily on the design, implementation, and management of the digital platform being used. OBJECTIVE: We sought to design and build a secure, privacy-preserving, validated, participant-centric digital health research platform (DHRP) to recruit and enroll participants, collect multimodal data, and engage participants from diverse backgrounds in the National Institutes of Health's (NIH) All of Us Research Program (AOU). AOU is an ongoing national, multiyear study aimed to build a research cohort of 1 million participants that reflects the diversity of the United States, including minority, health-disparate, and other populations underrepresented in biomedical research (UBR). METHODS: We collaborated with community members, health care provider organizations (HPOs), and NIH leadership to design, build, and validate a secure, feature-rich digital platform to facilitate multisite, hybrid, and remote study participation and multimodal data collection in AOU. Participants were recruited by in-person, print, and online digital campaigns. Participants securely accessed the DHRP via web and mobile apps, either independently or with research staff support. The participant-facing tool facilitated electronic informed consent (eConsent), multisource data collection (eg, surveys, genomic results, wearables, and electronic health records [EHRs]), and ongoing participant engagement. We also built tools for research staff to conduct remote participant support, study workflow management, participant tracking, data analytics, data harmonization, and data management. RESULTS: We built a secure, participant-centric DHRP with engaging functionality used to recruit, engage, and collect data from 705,719 diverse participants throughout the United States. As of April 2024, 87% (n=613,976) of the participants enrolled via the platform were from UBR groups, including racial and ethnic minorities (n=282,429, 46%), rural dwelling individuals (n=49,118, 8%), those over the age of 65 years (n=190,333, 31%), and individuals with low socioeconomic status (n=122,795, 20%). CONCLUSIONS: We built a participant-centric digital platform with tools to enable engagement with individuals from different racial, ethnic, and socioeconomic backgrounds and other UBR groups. This DHRP demonstrated successful use among diverse participants. These findings could be used as best practices for the effective use of digital platforms to build and sustain cohorts of various study designs and increase engagement with diverse populations in health research.

Humans↗

PubMed2004

Workflow improvement and impact of the new Beckman Coulter LH 1500 high throughput automated hematology workcell.

Our core laboratory at St. Joseph's Medical Center (SJMC) recently installed the new Beckman Coulter LH 1500 fully automated hematology workcell. Like many laboratories in North America and other parts of the world, we are under significant pressure to improve productivity, reduce costs, and improve service provision to our hospital and clinician bases. The difficulty of meeting this challenge is often further compounded by an increasing shortage of qualified technologist personnel. We recently reported on the significant productivity gains with the introduction of the LH 750 hematology system into the core laboratory, including a 23% reduction in slide review rate, automation of reticulocyte and nucleated red blood cell counts, and an 82% increase in the pass-through rate. Evolution to the next stage of automation has come much quicker than anticipated owing to a steep rise in workload and an ever increasing shortage of trained and experienced technologists to fill vacancies in the 24-hour core laboratory operation. We chose to upgrade our LH 750 analyzers to the newly introduced LH 1500 hematology workcell in order to cope with the increased workload and shortage of available experienced hematology technologists. The LH 1500 is a compact, high throughput, fully automated hematology workcell. It is designed to automate all manual process and handling steps associated with hematology analyses, including sample sorting, loading and unloading of cassettes, automatic rerun and reflex testing, sample storage, and tracking of samples. The LH 1500 is comprised of a minimum of 2, and up to 4, LH 750 instruments (with or without integrated slidemaker/stainers). The instruments are bidirectionally interfaced and linked to a fully automated conveyor track, specimen inlet and outlet stations, a sample stockyard, and a line controller. The configuration we chose at SJMC incorporates 1 LH 750 and 1 LH 755. We report that the upgrade to the LH 1500 was a relatively straightforward process and required no redesign of the core laboratory. Based on our evaluation, we estimate cost savings with the LH 1500 of between three quarters to 1 full-time equivalent, which eliminated or decreased many manual handling steps. In addition, this increase in efficiency has freed up our technologist resources to concentrate on the more difficult and clinically important samples. This is the first published report on the new LH 1500 hematology workcell. Our research indicates that the LH 1500 is the first hematology workcell to offer hematology laboratories the next level of automation and virtually eliminate preanalytical and postanalytical sample handling and sorting. For core laboratories and high-throughput hematology laboratories that face similar pressures to ours, the LH 1500 would appear as an attractive solution to maximize hematology productivity and improve service delivery.

Automation↗

PubMed2006

Avoiding prolonged waiting time during busy periods in the emergency department: Is there a role for the senior emergency physician in triage?

STUDY OBJECTIVE: Patient satisfaction at emergency departments can be improved by reductions in waiting time. Traditional methods require registration and triage before seeing the doctor with senior emergency physicians mainly engaged in treating serious cases. We examine a radical change in workflow pattern on waiting time by placing a senior emergency physician with the triage nurse and examining the impact of treating simple cases upfront with discharge on the waiting times for stretcher cases. METHODS: A senior emergency physician was placed with the triage nurse in the Department of Emergency Medicine at Alexandra Hospital during peak busy periods of patient attendance over a period of 2 months. Measures were made of waiting time (registration to doctor consult) of PACS 3 and PACS 2 (Patient Acvity Score) cases accordingly. RESULTS: Ten days were chosen for the changed workflow practice and 10 days for controls in which normal traditional working practice followed. On all days, there was the same number of medical staff. The average waiting time for walk-in patients (PACS 3) was 19 min on experimental days as compared with 35.5 min on control days, with 78% being seen within 30 min in the experimental group compared with 48% on control days (P < 0.05). The PACS 2 waiting time was also significantly decreased on experimental days (P < 0.05). CONCLUSIONS: Placing a senior emergency physician with the triage nurse reduced waiting times for walk-in cases. One third of attendances were treated and discharged quickly, allowing the consulting room and PACS 1/PACS 2 doctors to act more efficiently.

Cooperative Behavior↗

PubMedDate not supplied

Paperless medical records: reinventing the patient experience.

At North Shore Magnetic Imaging Center, the patient paper medical record system was becoming very cumbersome, and it served as a source of frustration for everyone involved: patients, technologists, radiologists, and staff members. The center's mapping of a typical patient experience indicated that, from the initial phone call scheduling an exam to a completed visit (claim processed and payment received), a record could be handled by as many as 20 sets of hands! In June 2002, the center's growth and a concern that patients were losing a one-on-one experience with the medical staff led to an evaluation of workflow processes existing at that time. The evaluation began with a survey of staff members, center management, radiologists, and referring physicians. Their responses indicated 3 common themes: stress, overload, and frustration over systems in place. Comments from the survey were grouped into 3 areas: Continue to Do, Stop Doing, and Start Doing. The Start Doing responses provided solid objectives. The center set out to establish a breakthrough project that included all stakeholders--patients, staff, management, and radiologists. The Reinvention Project had 2 primary goals: move to a paperless environment and increase the level of patient care. The project was divided into internal and external teams. The internal team, called the Reinvention Team, was responsible for the actual hands-on aspects of the process. There were numerous external teams; each had defined roles and specific outcomes for achievement. The external teams' responsibilities included implementing an Internet protocol telephone system; researching voice recognition; restructuring job descriptions, training manuals, and performance evaluations; and conducting a patient-centered focus group.

Hospital Shared Services↗

PubMed2006

User-centered design techniques for a computerised antibiotic decision support system in an intensive care unit.

OBJECTIVE: To explore the use of user-centered design techniques for developing the requirements for an antibiotic decision support system (DSS) in an intensive care unit (ICU). DESIGN AND METHODOLOGY: The setting was a 21-bed mixed medical/surgical adult ICU. This was an observational study with unstructured interviews and participatory design process. Models were constructed to demonstrate cultural, workflow, sequence/trigger events and other artefacts used to support antibiotic prescribing in the ICU. Using participatory design, a paper prototype was developed and case studies were used to simulate antibiotic prescribing for bacterial isolates. This information was used to design and pilot the decision support tool. RESULTS: The key users were identified as residents, registrars and the unit pharmacist. They identified the major requirements: ability to collate and print microbiology results, and to provide education and antibiotic advice for isolates. The final product was a real time microbiology browser and decision support tool for antibiotic prescribing (ADVISE). Uptake of the system was rapid with over 6000 encounters in the first 6 months. An audit of antibiotic use performed on all consecutive patients 6 months before and after introducing the DSS demonstrated a reduction in total and broad-spectrum antibiotics. CONCLUSION: Contextual design methodology in conjunction with participatory design was an effective method to design this antibiotic decision support tool. The process facilitated physician and pharmacist ownership of the system that resulted in immediate uptake and ongoing use.

Anti-Bacterial Agents↗

PubMedDate not supplied

Virtual automation.

Total laboratory automation (TLA) can be substituted in mid-size laboratories by a computer sample workflow control (virtual automation). Such a solution has been implemented in our laboratory using PSM, software developed in cooperation with Roche Diagnostics (Barcelona, Spain), to this purpose. This software is connected to the online analyzers and to the laboratory information system and is able to control and direct the samples working as an intermediate station. The only difference with TLA is the replacement of transport belts by personnel of the laboratory. The implementation of this virtual automation system has allowed us the achievement of the main advantages of TLA: workload increase (64%) with reduction in the cost per test (43%), significant reduction in the number of biochemistry primary tubes (from 8 to 2), less aliquoting (from 600 to 100 samples/day), automation of functional testing, drastic reduction of preanalytical errors (from 11.7 to 0.4% of the tubes) and better total response time for both inpatients (from up to 48 hours to up to 4 hours) and outpatients (from up to 10 days to up to 48 hours). As an additional advantage, virtual automation could be implemented without hardware investment and significant headcount reduction (15% in our lab).

Autoanalysis↗

PubMed2005

Human-centered design of a distributed knowledge management system.

Many healthcare technology projects fail due to the lack of consideration of human issues, such as workflow, organizational change, and usability, during the design and implementation stages of a project's development process. Even when human issues are considered, the consideration is typically on designing better user interfaces. We argue that human-centered computing goes beyond a better user interface: it should include considerations of users, functions and tasks that are fundamental to human-centered computing. From this perspective, we integrated a previously developed human-centered methodology with a Project Design Lifecycle, and we applied this integration in the design of a complex distributed knowledge management system for the Biomedical Engineer (BME) domain in the Mission Control Center at NASA Johnson Space Center. We analyzed this complex system, identified its problems, generated systems requirements, and provided specifications of a replacement prototype for effective organizational memory and knowledge management. We demonstrated the value provided by our human-centered approach and described the unique properties, structures, and processes discovered using this methodology and how they contributed in the design of the prototype.

Artificial Intelligence↗

PubMed2026

Artificial intelligence-assisted histopathological diagnosis of endocervical gastric-type adenocarcinoma: a multicenter model development and validation study.

Endocervical gastric-type adenocarcinoma (GAS) is one of the most aggressive subtypes of cervical cancer and is frequently underdiagnosed due to morphological ambiguity, leading to delayed diagnosis. Despite the availability of molecular and genomic assays, their high cost, complexity, and limited reproducibility restrict clinical use. This study therefore proposes a highly sensitive artificial intelligence (AI)-assisted diagnostic system for GAS based exclusively on H&E-stained histopathological images. We included 309 slides from 96 GAS cases collected at Peking University Third Hospital from January 2018 to January 2025, representing the largest GAS cohort reported to date for AI research. In addition, we incorporated other morphologically analogous diseases, encompassing a total of 1,320 slides sourced from four categories: normal cervical mucosa (NORM), benign endocervical lesion entities (BELE), HPV-associated adenocarcinoma (HPVA), and endometrioid carcinoma with mucinous differentiation (ECMD). We developed GASPath, based on a novel multiple instance learning framework that efficiently captures fine-grained morphological variations from H&E-stained images. Beyond internal validation, GASPath was evaluated across 12 independent retrospective cohorts and further subjected to large-scale real-world validation on more than 7,000 samples from March 2024 to April 2025. Across three stages, GASPath demonstrated high performance. In internal validation (Stage I), it achieved an accuracy of 0.980 (95% CI 0.977-0.983) and an ROC-AUC of 0.995 (95% CI 0.994-0.997). In external validation (Stage II), the sensitivity reached 0.902 and improved to 0.968 with proposed strategies. For biopsy samples, GASPath achieved an ROC-AUC of 0.990 (95% CI 0.984-0.997). In large-scale real-world deployment (Stage III, n = 7,056), GASPath achieved a balanced accuracy of 0.953, with 100% sensitivity for GAS (45/45 cases correctly identified). The heatmaps highlight morphological features of GAS that are easily underestimated, such as irregular, angulated glands, subtle loss of nuclear polarity, and mild cytologic atypia, which show substantial morphological overlap with other diagnostic categories. GASPath enables high-sensitivity detection of GAS in routine H&E-stained slides, obviating the need for extensive auxiliary testing while preventing underdiagnosis and misdiagnosis. This advancement addresses a critical gap by streamlining diagnostic workflows without compromising accuracy. Its implementation could enable cost-effective, scalable AI-assisted diagnostics, potentially transforming the early detection and management of this aggressive cancer subtype.

Female↗

PubMed2006

Implementation of a computerized physician order entry system of medications at the University Health Network--physicians' perspectives on the critical issues.

There are many reasons why most hospitals have not adopted physician order entry systems for medications. It is a costly endeavour (Kuperman and Gibson 2003) that can cause major disruptions to workflow for physicians, pharmacists and nurses. Yet, the technology can reduce medication errors, especially with sophisticated decision support. We have presented many of the lessons learned from our successful implementation experience. To date, over 90% of medication orders are entered by physicians. The technology must be ready for the implementation. System issues such as errors, slowness and freezing give ready opportunity for critics who will claim the system is just not ready for real-time. Through rigorous testing, we were able to avoid issues previously seen in our pilot study. Usability testing with end-users was also critical in both guiding decision-making as well as validating that the system was ready for implementation. Proper training and support were also necessary. To ensure ready adoption, decision support was optimized to reduce the volume of less important alerts. Most importantly, we found that active physician involvement at multiple levels was key. This ensured that physicians understood from a high-level perspective that this change was necessary. Planning for specific implementation details had the benefit of input from physicians working in the area. Day-to-day issues of our residents and staff were also addressed promptly.

Attitude of Health Personnel↗

PubMed2026

Scalable, open-access and multidisciplinary data integration pipeline for climate-sensitive diseases.

Climate-sensitive infectious diseases pose an important challenge for human, animal and environmental health and it has been estimated that over half of known human pathogenic diseases can be aggravated by climate change. While climatic and weather conditions are important drivers of transmission of vector-borne diseases, socio-economic, behavioural, and land-use factors as well as the interactions among them impact transmission dynamics. Analysis of drivers of climate-sensitive diseases require rapid integration of interdisciplinary data to be jointly analysed with epidemiological (including genomic and clinical) data. Current tools for the integration of multiple data sources are often limited to one data type or rely on proprietary data and software. To address this gap, we develop a scalable and open-access pipeline for the integration of multiple spatio-temporal datasets that requires only the declaration of the country and temporal range and resolution of the study. The tool is locally deployable and can easily be integrated into existing climate-disease-modelling applications. We demonstrate the utility of the tool for dengue modelling in Vietnam where epidemiological data are legally required to remain local. We include a pipeline for bias correction of climate data to enhance their quality for downstream modelling tasks. The Dengue Advanced Readiness Tools-Pipeline empowers users by simplifying complex download, correction, and aggregation steps, fostering data-driven discovery of relationships between infectious diseases and their drivers in space and time, and enhancing reproducibility in research. Additional modules and datasets can be added to the existing ones to make the pipeline extendable to use cases other than the ones presented here.

automated workflows↗