Journal of Medical Systems最新文献

Anesthesiologists have a significant responsibility to provide care at all hours of the day, including nights, weekends, and holidays. This call burden carries a significant lifestyle constraint that can impact relationships, affect provider wellbeing, and has been associated with provider burnout. This quality improvement study analyzes the effects of a dynamic call marketplace, which allows anesthesiologists to specify how much call they would like to take across a spectrum of hypothetical compensation levels, from very low to very high. The system then determines the market equilibrium price such that every anesthesiologist gets exactly the amount of desired call. A retrospective analysis compared percentage participation in adjusting call burden both pre- and post-implementation of a dynamic marketplace during the years of 2017 to 2023. Additionally, a 2023 post-implementation survey was sent out assessing various aspects of anesthesiologist perception of the new system including work-life balance and job satisfaction. The dynamic call marketplace in this study enabled a more effective platform for adjusting call levels, as there was a statistically significant increase in the percentage of anesthesiologists participating in call exchanged during post- compared to pre-implementation (p < 0.0001). The satisfaction survey suggested agreement among anesthesiologists that the dynamic call marketplace positively affected professional satisfaction and work-life balance. Further, the level of agreement with these statements was most prevalent among middle career stage anesthesiologists (11-20 years as attending physician). The present system may target elements with the capacity to increase satisfaction, particularly among physicians most at risk of burnout within the anesthesia workforce.

This retrospective study assessed the effectiveness and impact of implementing a Modified Early Warning System (MEWS) and Rapid Response Team (RRT) for inpatients admitted to the general ward (GW) of a medical center. This study included all inpatients who stayed in GWs from Jan. 2017 to Feb. 2022. We divided inpatients into GW_non-MEWS and GW_MEWS groups according to MEWS and RRT implementation in Aug. 2019. The primary outcome, unexpected deterioration, was defined by unplanned admission to intensive care units. We defined the detection performance and effectiveness of MEWS according to if a warning occurred within 24 h before the unplanned ICU admission. There were 129,039 inpatients included in this study, comprising 58,106 GW_non-MEWS and 71,023 GW_MEWS. The numbers of inpatients who underwent an unplanned ICU admission in GW_non-MEWS and GW_MEWS were 488 (.84%) and 468 (.66%), respectively, indicating that the implementation significantly reduced unexpected deterioration (p < .0001). Besides, 1,551,525 times MEWS assessments were executed for the GW_MEWS. The sensitivity, specificity, positive predicted value, and negative predicted value of the MEWS were 29.9%, 98.7%, 7.09%, and 99.76%, respectively. A total of 1,568 warning signs accurately occurred within the 24 h before an unplanned ICU admission. Among them, 428 (27.3%) met the criteria for automatically calling RRT, and 1,140 signs necessitated the nursing staff to decide if they needed to call RRT. Implementing MEWS and RRT increases nursing staff's monitoring and interventions and reduces unplanned ICU admissions.

In healthcare professions, soft skills contribute to critical thinking, decision-making, and patient-centered care. While important to the delivery of high-quality medical care, soft skills are often underemphasized during healthcare training in low-and-middle-income countries. Despite South Asia's large population, the efficacy and viability of a digital soft skills curriculum for South Asian healthcare practitioners has not been studied to date. We hypothesized that a web-based, multilingual, soft skills course could aid the understanding and application of soft skills to improve healthcare practitioner knowledge, confidence, attitudes, and intent-to-change clinical practice.In September 2019 a needs assessment observing soft skills practices was conducted in several Indian states. We developed a communication-focused soft skills curriculum that comprised seven 10-minute video lectures, recorded in spoken English and Hindi. Participants consisted of any practicing healthcare professionals and trainees in select South Asian countries age 18 and over. Participant knowledge, confidence, attitudes, and intent-to-change clinical practice were evaluated using pre- and post-course tests and surveys. Statistical analyses were performed using STATA and SPSS.From July 26, 2021 to September 26, 2021, 5750 registered and attempted the course, 2628 unique participants completed the pre-test, and 1566 unique participants completed the post-test. Participants demonstrated small but statistically significant gains in confidence (𝑝<0.001), attitudes toward course topics relevance (𝑝<0.001), and intent-to-change clinical practice (𝑝<0.001). There was no statistically significant gain in knowledge. A digital soft-skills massive open online course for healthcare practitioners in South Asia could serve as a viable approach to improve the quality of soft skills training in low-to-middle income countries.

Intraoperative cardiopulmonary variables are well-known predictors of postoperative pulmonary complications (PPC), traditionally quantified by median values over the duration of surgery. However, it is unknown whether cardiopulmonary instability, or wider intra-operative variability of the same metrics, is distinctly associated with PPC risk and severity. We leveraged a retrospective cohort of adults (n = 1202) undergoing major non-cardiothoracic surgery. We used multivariable logistic regression to evaluate the association of two outcomes (1)moderate-or-severe PPC and (2)any PPC with two sets of exposure variables- (a)variability of cardiopulmonary metrics (inter-quartile range, IQR) and (b)median intraoperative cardiopulmonary metrics. We compared predictive ability (receiver operating curve analysis, ROC) and parsimony (information criteria) of three models evaluating different aspects of the intra-operative cardiopulmonary metrics: Median-based: Median cardiopulmonary metrics alone, Variability-based: IQR of cardiopulmonary metrics alone, and Combined: Medians and IQR. Models controlled for peri-operative/surgical factors, demographics, and comorbidities. PPC occurred in 400(33%) of patients, and 91(8%) experienced moderate-or-severe PPC. Variability in multiple intra-operative cardiopulmonary metrics was independently associated with risk of moderate-or-severe, but not any, PPC. For moderate-or-severe PPC, the best-fit predictive model was the Variability-based model by both information criteria and ROC analysis (area under the curve, AUC_{Variability-based} = 0.74 vs AUC_Median-based = 0.65, p = 0.0015; AUC_{Variability-based} = 0.74 vs AUC_Combined = 0.68, p = 0.012). For any PPC, the Median-based model yielded the best fit by information criteria. Predictive accuracy was marginally but not significantly higher for the Combined model (AUC_Combined = 0.661) than for the Median-based (AUC_Median-based = 0.657, p = 0.60) or Variability-based (AUC_{Variability-based} = 0.649, p = 0.29) models. Variability of cardiopulmonary metrics, distinct from median intra-operative values, is an important predictor of moderate-or-severe PPC.

Although magnetic resonance imaging (MRI) data of patients with multiple myeloma (MM) are used to predict prognosis, few reports have applied artificial intelligence (AI) techniques for this purpose. We aimed to analyze whole-body diffusion-weighted MRI data using three-dimensional (3D) convolutional neural networks (CNNs) and Gradient-weighted Class Activation Mapping (Grad-CAM), an explainable AI, to predict prognosis and explore the factors involved in prediction. We retrospectively analyzed the MRI data of a total of 142 patients with MM obtained from two medical centers. We defined the occurrence of progressive disease after MRI evaluation within 12 months as a poor prognosis and constructed a 3D CNN-based deep learning model to predict prognosis. Images from 111 cases were used as the training and internal validation data; images from 31 cases were used as the external validation data. Internal validation of the AI model with stratified 5-fold cross-validation resulted in a significant difference in progression-free survival (PFS) between good and poor prognostic cases (2-year PFS, 91.2% versus [vs.] 61.1%, P = 0.0002). The AI model clearly stratified good and poor prognostic cases in the external validation cohort (2-year PFS, 92.9% vs. 55.6%, P = 0.004), with an area under the receiver operating characteristic curve of 0.804. According to Grad-CAM, the MRI signals of the spleen and bones of the vertebrae and pelvis contributed to prognosis prediction. This study is the first to show that image analysis of whole-body MRI using a 3D CNN without any other clinical data is effective in predicting the prognosis of patients with MM.

Schizophrenia is a serious chronic mental disorder that significantly affects daily life. Electroencephalography (EEG), a method used to measure mental activities in the brain, is among the techniques employed in the diagnosis of schizophrenia. The symptoms of the disease typically begin in childhood and become more pronounced as one grows older. However, it can be managed with specific treatments. Computer-aided methods can be used to achieve an early diagnosis of this illness. In this study, various machine learning algorithms and the emerging technology of quantum-based machine learning algorithm were used to detect schizophrenia using EEG signals. The principal component analysis (PCA) method was applied to process the obtained data in quantum systems. The data, which were reduced in dimensionality, were transformed into qubit form using various feature maps and provided as input to the Quantum Support Vector Machine (QSVM) algorithm. Thus, the QSVM algorithm was applied using different qubit numbers and different circuits in addition to classical machine learning algorithms. All analyses were conducted in the simulator environment of the IBM Quantum Platform. In the classification of this EEG dataset, it is evident that the QSVM algorithm demonstrated superior performance with a 100% success rate when using Pauli X and Pauli Z feature maps. This study serves as proof that quantum machine learning algorithms can be effectively utilized in the field of healthcare.

Background: The Virtual Reality Headset (VRH) is a device aiming at improving patient's comfort by reducing pain and anxiety during medical interventions. Its interest during cardiac implantable electronic devices (CIED) implant procedures has not been studied.

Methods: We randomized consecutive patients admitted for pacemaker or Implantable Cardioverter Defibrillator (ICD) at our center to either standard analgesia care (STD-Group), or to VRH (VRH-Group). Patients in the STD-Group received intra-venous paracetamol (1 g) 60 min before the procedure, and local anesthesia was performed with lidocaine. For patients of the VRH-Group, VRH was used on top of standard care. We monitored patients' pain and anxiety using numeric rating scales (from 0 to 10) at the time of sub-cutaneous pocket creation, and during deep axillary vein puncture. Patient comfort during the procedure was assessed using a detailed questionnaire. Morphine consumption was also assessed.

Results: We randomized 61 patients to STD-Group (n = 31) or VRH-Group (n = 30). Pain and anxiety were lower in the VRH-Group during deep venous puncture (3.0 ± 2.0 vs. 4.8 ± 2.2, p = 0.002 and 2.4 ± 2.2 vs. 4.1 ± 2.4, p = 0.006) but not during pocket creation (p = 0.58 and p = 0.5). Morphine consumption was lower in the VRH-Group (1.6 ± 0.7 vs. 2.1 ± 1.1 mg; p = 0.041). Patients' overall comfort during procedure was similar in both groups.

Conclusion: VRH use improved pain and anxiety control during deep venous puncture compared to standard analgesia care, and allowed morphine consumption reduction. However, pain and anxiety were similar at the time of sub-cutaneous pocket creation.

Clinical abbreviation disambiguation is a crucial task in the biomedical domain, as the accurate identification of the intended meanings or expansions of abbreviations in clinical texts is vital for medical information retrieval and analysis. Existing approaches have shown promising results, but challenges such as limited instances and ambiguous interpretations persist. In this paper, we propose an approach to address these challenges and enhance the performance of clinical abbreviation disambiguation. Our objective is to leverage the power of Large Language Models (LLMs) and employ a Generative Model (GM) to augment the dataset with contextually relevant instances, enabling more accurate disambiguation across diverse clinical contexts. We integrate the contextual understanding of LLMs, represented by BlueBERT and Transformers, with data augmentation using a Generative Model, called Biomedical Generative Pre-trained Transformer (BIOGPT), that is pretrained on an extensive corpus of biomedical literature to capture the intricacies of medical terminology and context. By providing the BIOGPT with relevant medical terms and sense information, we generate diverse instances of clinical text that accurately represent the intended meanings of abbreviations. We evaluate our approach on the widely recognized CASI dataset, carefully partitioned into training, validation, and test sets. The incorporation of data augmentation with the GM improves the model's performance, particularly for senses with limited instances, effectively addressing dataset imbalance and challenges posed by similar concepts. The results demonstrate the efficacy of our proposed method, showcasing the significance of LLMs and generative techniques in clinical abbreviation disambiguation. Our model achieves a good accuracy on the test set, outperforming previous methods.

Introduction: ChatGPT, a recently released chatbot from OpenAI, has found applications in various aspects of life, including academic research. This study investigated the knowledge, perceptions, and attitudes of researchers towards using ChatGPT and other chatbots in academic research.

Methods: A pre-designed, self-administered survey using Google Forms was employed to conduct the study. The questionnaire assessed participants' knowledge of ChatGPT and other chatbots, their awareness of current chatbot and artificial intelligence (AI) applications, and their attitudes towards ChatGPT and its potential research uses.

Results: Two hundred researchers participated in the survey. A majority were female (57.5%), and over two-thirds belonged to the medical field (68%). While 67% had heard of ChatGPT, only 11.5% had employed it in their research, primarily for rephrasing paragraphs and finding references. Interestingly, over one-third supported the notion of listing ChatGPT as an author in scientific publications. Concerns emerged regarding AI's potential to automate researcher tasks, particularly in language editing, statistics, and data analysis. Additionally, roughly half expressed ethical concerns about using AI applications in scientific research.

Conclusion: The increasing use of chatbots in academic research necessitates thoughtful regulation that balances potential benefits with inherent limitations and potential risks. Chatbots should not be considered authors of scientific publications but rather assistants to researchers during manuscript preparation and review. Researchers should be equipped with proper training to utilize chatbots and other AI tools effectively and ethically.

Precise neurosurgical guidance is critical for successful brain surgeries and plays a vital role in all phases of image-guided neurosurgery (IGN). Neuronavigation software enables real-time tracking of surgical tools, ensuring their presentation with high precision in relation to a virtual patient model. Therefore, this work focuses on the development of a novel multimodal IGN system, leveraging deep learning and explainable AI to enhance brain tumor surgery outcomes. The study establishes the clinical and technical requirements of the system for brain tumor surgeries. NeuroIGN adopts a modular architecture, including brain tumor segmentation, patient registration, and explainable output prediction, and integrates open-source packages into an interactive neuronavigational display. The NeuroIGN system components underwent validation and evaluation in both laboratory and simulated operating room (OR) settings. Experimental results demonstrated its accuracy in tumor segmentation and the success of ExplainAI in increasing the trust of medical professionals in deep learning. The proposed system was successfully assembled and set up within 11 min in a pre-clinical OR setting with a tracking accuracy of 0.5 (± 0.1) mm. NeuroIGN was also evaluated as highly useful, with a high frame rate (19 FPS) and real-time ultrasound imaging capabilities. In conclusion, this paper describes not only the development of an open-source multimodal IGN system but also demonstrates the innovative application of deep learning and explainable AI algorithms in enhancing neuronavigation for brain tumor surgeries. By seamlessly integrating pre- and intra-operative patient image data with cutting-edge interventional devices, our experiments underscore the potential for deep learning models to improve the surgical treatment of brain tumors and long-term post-operative outcomes.