Deep Learning-Based Localization and Detection of Malpositioned Endotracheal Tube on Portable Supine Chest Radiographs in Intensive and Emergency Medicine: A Multicenter Retrospective Study.

OBJECTIVES: We aimed to develop a computer-aided detection (CAD) system to localize and detect the malposition of endotracheal tubes (ETTs) on portable supine chest radiographs (CXRs). DESIGN: This was a retrospective diagnostic study. DeepLabv3+ with ResNeSt50 backbone and DenseNet121 served as the model architecture for segmentation and classification tasks, respectively. SETTING: Multicenter study. PATIENTS: For the training dataset, images meeting the following inclusion criteria were included: 1) patient age greater than or equal to 20 years; 2) portable supine CXR; 3) examination in emergency departments or ICUs; and 4) examination between 2015 and 2019 at National Taiwan University Hospital (NTUH) (NTUH-1519 dataset: 5,767 images). The derived CAD system was tested on images from chronologically (examination during 2020 at NTUH, NTUH-20 dataset: 955 images) or geographically (examination between 2015 and 2020 at NTUH Yunlin Branch [YB], NTUH-YB dataset: 656 images) different datasets. All CXRs were annotated with pixel-level labels of ETT and with image-level labels of ETT presence and malposition. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: For the segmentation model, the Dice coefficients indicated that ETT would be delineated accurately (NTUH-20: 0.854; 95% CI, 0.824-0.881 and NTUH-YB: 0.839; 95% CI, 0.820-0.857). For the classification model, the presence of ETT could be accurately detected with high accuracy (area under the receiver operating characteristic curve [AUC]: NTUH-20, 1.000; 95% CI, 0.999-1.000 and NTUH-YB: 0.994; 95% CI, 0.984-1.000). Furthermore, among those images with ETT, ETT malposition could be detected with high accuracy (AUC: NTUH-20, 0.847; 95% CI, 0.671-0.980 and NTUH-YB, 0.734; 95% CI, 0.630-0.833), especially for endobronchial intubation (AUC: NTUH-20, 0.991; 95% CI, 0.969-1.000 and NTUH-YB, 0.966; 95% CI, 0.933-0.991). CONCLUSIONS: The derived CAD system could localize ETT and detect ETT malposition with excellent performance, especially for endobronchial intubation, and with favorable potential for external generalizability.

A Delphi Method for Development of a Barrett's Esophagus Question Prompt List as a Communication Tool for Optimal Patient-physician Communication.

BACKGROUND METHODS: The question prompt list content was derived through a modified Delphi process consisting of 3 rounds. In round 1, experts provided 5 answers to the prompts "What general questions should patients ask when given a new diagnosis of Barrett's esophagus" and "What questions do I not hear patients asking, but given my expertise, I believe they should be asking?" Questions were reviewed and categorized into themes. In round 2, experts rated questions on a 5-point Likert scale. In round 3, experts rerated questions modified or reduced after the previous rounds. Only questions rated as "essential" or "important" were included in Barrett's esophagus question prompt list (BE-QPL). To improve usability, questions were reduced to minimize redundancy and simplified to use language at an eighth-grade level (Fig. 1). RESULTS: Twenty-one esophageal medical and surgical experts participated in both rounds (91% males; median age 52 years). The expert panel comprised of 33% esophagologists, 24% foregut surgeons, and 24% advanced endoscopists, with a median of 15 years in clinical practice. Most (81%), worked in an academic tertiary referral hospital. In this 3-round Delphi technique, 220 questions were proposed in round 1, 122 (55.5%) were accepted into the BE-QPL and reduced down to 76 questions (round 2), and 67 questions (round 3). These 67 questions reached a Flesch Reading Ease of 68.8, interpreted as easily understood by 13 to 15 years olds. CONCLUSIONS: With multidisciplinary input, we have developed a physician-derived BE-QPL to optimize patient-physician communication. Future directions will seek patient feedback to distill the questions further to a smaller number and then assess their usability.

Outcome prediction of cardiac arrest with automatically computed gray-white matter ratio on computed tomography images.

BACKGROUND: This study aimed to develop an automated method to measure the gray-white matter ratio (GWR) from brain computed tomography (CT) scans of patients with out-of-hospital cardiac arrest (OHCA) and assess its significance in predicting early-stage neurological outcomes. METHODS: Patients with OHCA who underwent brain CT imaging within 12 h of return of spontaneous circulation were enrolled in this retrospective study. The primary outcome endpoint measure was a favorable neurological outcome, defined as cerebral performance category 1 or 2 at hospital discharge. We proposed an automated method comprising image registration, K-means segmentation, segmentation refinement, and GWR calculation to measure the GWR for each CT scan. The K-means segmentation and segmentation refinement was employed to refine the segmentations within regions of interest (ROIs), consequently enhancing GWR calculation accuracy through more precise segmentations. RESULTS: Overall, 443 patients were divided into derivation N=265, 60% and validation N=178, 40% sets, based on age and sex. The ROI Hounsfield unit values derived from the automated method showed a strong correlation with those obtained from the manual method. Regarding outcome prediction, the automated method significantly outperformed the manual method in GWR calculation (AUC 0.79 vs. 0.70) across the entire dataset. The automated method also demonstrated superior performance across sensitivity, specificity, and positive and negative predictive values using the cutoff value determined from the derivation set. Moreover, GWR was an independent predictor of outcomes in logistic regression analysis. Incorporating the GWR with other clinical and resuscitation variables significantly enhanced the performance of prediction models compared to those without the GWR. CONCLUSIONS: Automated measurement of the GWR from non-contrast brain CT images offers valuable insights for predicting neurological outcomes during the early post-cardiac arrest period.

Optimal methods of vitamin D supplementation to prevent acute respiratory infections: a systematic review, dose-response and pairwise meta-analysis of randomized controlled trials.

BACKGROUND: Vitamin D supplementation may prevent acute respiratory infections (ARIs). This study aimed to identify the optimal methods of vitamin D supplementation. METHODS: PubMed, Embase, Cochrane Central Register of Controlled Trials, Web of Science, and the ClinicalTrials.gov registry were searched from database inception through July 13, 2023. Randomized-controlled trials (RCTs) were included. Data were pooled using random-effects model. The primary outcome was the proportion of participants with one or more ARIs. RESULTS: The analysis included 43 RCTs with 49320 participants. Forty RCTs were considered to be at low risk for bias. The main pairwise meta-analysis indicated there were no significant preventive effects of vitamin D supplementation against ARIs (risk ratio [RR]: 0.99, 95% confidence interval [CI]: 0.97 to 1.01, I2 = 49.6%). The subgroup dose-response meta-analysis indicated that the optimal vitamin D supplementation doses ranged between 400-1200 IU/day for both summer-sparing and winter-dominant subgroups. The subgroup pairwise meta-analysis also revealed significant preventive effects of vitamin D supplementation in subgroups of daily dosing (RR: 0.92, 95% CI: 0.85 to 0.99, I2 = 55.7%, number needed to treat [NNT]: 36), trials duration < 4 months (RR: 0.81, 95% CI: 0.67 to 0.97, I2 = 48.8%, NNT: 16), summer-sparing seasons (RR: 0.85, 95% CI: 0.74 to 0.98, I2 = 55.8%, NNT: 26), and winter-dominant seasons (RR: 0.79, 95% CI: 0.71 to 0.89, I2 = 9.7%, NNT: 10). CONCLUSION: Vitamin D supplementation may slightly prevent ARIs when taken daily at doses between 400 and 1200 IU/d during spring, autumn, or winter, which should be further examined in future clinical trials.

Large language model application in emergency medicine and critical care.

In the rapidly evolving healthcare landscape, artificial intelligence (AI), particularly the large language models (LLMs), like OpenAI's Chat Generative Pretrained Transformer (ChatGPT), has shown transformative potential in emergency medicine and critical care. This review article highlights the advancement and applications of ChatGPT, from diagnostic assistance to clinical documentation and patient communication, demonstrating its ability to perform comparably to human professionals in medical examinations. ChatGPT could assist clinical decision-making and medication selection in critical care, showcasing its potential to optimize patient care management. However, integrating LLMs into healthcare raises legal, ethical, and privacy concerns, including data protection and the necessity for informed consent. Finally, we addressed the challenges related to the accuracy of LLMs, such as the risk of providing incorrect medical advice. These concerns underscore the importance of ongoing research and regulation to ensure their ethical and practical use in healthcare.

C-Phycocyanin Attenuates Noise-Induced Cochlear Synaptopathy via the Inhibition of Oxidative Stress and Intercellular Adhesion Molecule-1 in the Cochlea.

The synapses between inner hair cells (IHCs) and spiral ganglion neurons (SGNs) are the most vulnerable structures in the noise-exposed cochlea. Cochlear synaptopathy results from the disruption of these synapses following noise exposure and is considered the main cause of poor speech understanding in noisy environments, even when audiogram results are normal. Cochlear synaptopathy leads to the degeneration of SGNs if damaged IHC-SGN synapses are not promptly recovered. Oxidative stress plays a central role in the pathogenesis of cochlear synaptopathy. C-Phycocyanin (C-PC) has antioxidant and anti-inflammatory activities and is widely utilized in the food and drug industry. However, the effect of the C-PC on noise-induced cochlear damage is unknown. We first investigated the therapeutic effect of C-PC on noise-induced cochlear synaptopathy. In vitro experiments revealed that C-PC reduced the H2O2-induced generation of reactive oxygen species in HEI-OC1 auditory cells. H2O2-induced cytotoxicity in HEI-OC1 cells was reduced with C-PC treatment. After white noise exposure for 3 h at a sound pressure of 118 dB, the guinea pigs intratympanically administered 5 µg/mL C-PC exhibited greater wave I amplitudes in the auditory brainstem response, more IHC synaptic ribbons and more IHC-SGN synapses according to microscopic analysis than the saline-treated guinea pigs. Furthermore, the group treated with C-PC had less intense 4-hydroxynonenal and intercellular adhesion molecule-1 staining in the cochlea compared with the saline group. Our results suggest that C-PC improves cochlear synaptopathy by inhibiting noise-induced oxidative stress and the inflammatory response in the cochlea.

Ultrasound-Mediated Lysozyme Microbubbles Targeting NOX4 Knockdown Alleviate Cisplatin-Exposed Cochlear Hair Cell Ototoxicity.

The nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 4 (NOX4) protein plays an essential role in the cisplatin (CDDP)-induced generation of reactive oxygen species (ROS). In this study, we evaluated the suitability of ultrasound-mediated lysozyme microbubble (USMB) cavitation to enhance NOX4 siRNA transfection in vitro and ex vivo. Lysozyme-shelled microbubbles (LyzMBs) were constructed and designed for siNOX4 loading as siNOX4/LyzMBs. We investigated different siNOX4-based cell transfection approaches, including naked siNOX4, LyzMB-mixed siNOX4, and siNOX4-loaded LyzMBs, and compared their silencing effects in CDDP-treated HEI-OC1 cells and mouse organ of Corti explants. Transfection efficiencies were evaluated by quantifying the cellular uptake of cyanine 3 (Cy3) fluorescein-labeled siRNA. In vitro experiments showed that the high transfection efficacy (48.18%) of siNOX4 to HEI-OC1 cells mediated by US and siNOX4-loaded LyzMBs significantly inhibited CDDP-induced ROS generation to almost the basal level. The ex vivo CDDP-treated organ of Corti explants of mice showed an even more robust silencing effect of the NOX4 gene in the siNOX4/LyzMB groups treated with US sonication than without US sonication, with a marked abolition of CDDP-induced ROS generation and cytotoxicity. Loading of siNOX4 on LyzMBs can stabilize siNOX4 and prevent its degradation, thereby enhancing the transfection and silencing effects when combined with US sonication. This USMB-derived therapy modality for alleviating CDDP-induced ototoxicity may be suitable for future clinical applications.

Artificial Intelligence-enabled Chest X-ray Classifies Osteoporosis and Identifies Mortality Risk.

A deep learning model was developed to identify osteoporosis from chest X-ray (CXR) features with high accuracy in internal and external validation. It has significant prognostic implications, identifying individuals at higher risk of all-cause mortality. This Artificial Intelligence (AI)-enabled CXR strategy may function as an early detection screening tool for osteoporosis. The aim of this study was to develop a deep learning model (DLM) to identify osteoporosis via CXR features and investigate the performance and clinical implications. This study collected 48,353 CXRs with the corresponding T score according to Dual energy X-ray Absorptiometry (DXA) from the academic medical center. Among these, 35,633 CXRs were used to identify CXR- Osteoporosis (CXR-OP). Another 12,720 CXRs were used to validate the performance, which was evaluated by the area under the receiver operating characteristic curve (AUC). Furthermore, CXR-OP was tested to assess the long-term risks of mortality, which were evaluated by Kaplan‒Meier survival analysis and the Cox proportional hazards model. The DLM utilizing CXR achieved AUCs of 0.930 and 0.892 during internal and external validation, respectively. The group that underwent DXA with CXR-OP had a higher risk of all-cause mortality (hazard ratio [HR] 2.59, 95% CI: 1.83-3.67), and those classified as CXR-OP in the group without DXA also had higher all-cause mortality (HR: 1.67, 95% CI: 1.61-1.72) in the internal validation set. The external validation set produced similar results. Our DLM uses CXRs for early detection of osteoporosis, aiding physicians to identify those at risk. It has significant prognostic implications, improving life quality and reducing mortality. AI-enabled CXR strategy may serve as a screening tool.

Artificial Intelligence-Enabled Electrocardiography Predicts Future Pacemaker Implantation and Adverse Cardiovascular Events.

Medical advances prolonging life have led to more permanent pacemaker implants. When pacemaker implantation (PMI) is commonly caused by sick sinus syndrome or conduction disorders, predicting PMI is challenging, as patients often experience related symptoms. This study was designed to create a deep learning model (DLM) for predicting future PMI from ECG data and assess its ability to predict future cardiovascular events. In this study, a DLM was trained on a dataset of 158,471 ECGs from 42,903 academic medical center patients, with additional validation involving 25,640 medical center patients and 26,538 community hospital patients. Primary analysis focused on predicting PMI within 90 days, while all-cause mortality, cardiovascular disease (CVD) mortality, and the development of various cardiovascular conditions were addressed with secondary analysis. The study's raw ECG DLM achieved area under the curve (AUC) values of 0.870, 0.878, and 0.883 for PMI prediction within 30, 60, and 90 days, respectively, along with sensitivities exceeding 82.0% and specificities over 81.9% in the internal validation. Significant ECG features included the PR interval, corrected QT interval, heart rate, QRS duration, P-wave axis, T-wave axis, and QRS complex axis. The AI-predicted PMI group had higher risks of PMI after 90 days (hazard ratio [HR]: 7.49, 95% CI: 5.40-10.39), all-cause mortality (HR: 1.91, 95% CI: 1.74-2.10), CVD mortality (HR: 3.53, 95% CI: 2.73-4.57), and new-onset adverse cardiovascular events. External validation confirmed the model's accuracy. Through ECG analyses, our AI DLM can alert clinicians and patients to the possibility of future PMI and related mortality and cardiovascular risks, aiding in timely patient intervention.

Deep Learning Can Predict Bevacizumab Therapeutic Effect and Microsatellite Instability Directly from Histology in Epithelial Ovarian Cancer.

Epithelial ovarian cancer (EOC) remains a significant cause of mortality among gynecologic cancers, with the majority of cases being diagnosed at an advanced stage. Before targeted therapies were available, EOC treatment relied largely on debulking surgery and platinum-based chemotherapy. Vascular endothelial growth factors have been identified as inducing tumor angiogenesis. According to several clinical trials, anti-vascular endothelial growth factor-targeted therapy with bevacizumab was effective in all phases of EOC treatment. However, there are currently no biomarkers accessible for regular therapeutic use despite the importance of patient selection. Microsatellite instability (MSI), caused by a deficiency of the DNA mismatch repair system, is a molecular abnormality observed in EOC associated with Lynch syndrome. Recent evidence suggests that angiogenesis and MSI are interconnected. Developing predictive biomarkers, which enable the selection of patients who might benefit from bevacizumab-targeted therapy or immunotherapy, is critical for realizing personalized precision medicine. In this study, we developed 2 improved deep learning methods that eliminate the need for laborious detailed image-wise annotations by pathologists and compared them with 3 state-of-the-art methods to not only predict the efficacy of bevacizumab in patients with EOC using mismatch repair protein immunostained tissue microarrays but also predict MSI status directly from histopathologic images. In prediction of therapeutic outcomes, the 2 proposed methods achieved excellent performance by obtaining the highest mean sensitivity and specificity score using MSH2 or MSH6 markers and outperformed 3 state-of-the-art deep learning methods. Moreover, both statistical analysis results, using Cox proportional hazards model analysis and Kaplan-Meier progression-free survival analysis, confirm that the 2 proposed methods successfully differentiate patients with positive therapeutic effects and lower cancer recurrence rates from patients experiencing disease progression after treatment (P < .01). In prediction of MSI status directly from histopathology images, our proposed method also achieved a decent performance in terms of mean sensitivity and specificity score even for imbalanced data sets for both internal validation using tissue microarrays from the local hospital and external validation using whole section slides from The Cancer Genome Atlas archive.

Design and rationale of re-energize fontan: Randomized exercise intervention designed to maximize fitness in fontan patients.

In this manuscript, we describe the design and rationale of a randomized controlled trial in pediatric Fontan patients to test the hypothesis that a live-video-supervised exercise (aerobic+resistance) intervention will improve cardiac and physical capacity; muscle mass, strength, and function; and endothelial function. Survival of children with single ventricles beyond the neonatal period has increased dramatically with the staged Fontan palliation. Yet, long-term morbidity remains high. By age 40, 50% of Fontan patients will have died or undergone heart transplantation. Factors that contribute to onset and progression of heart failure in Fontan patients remain incompletely understood. However, it is established that Fontan patients have poor exercise capacity which is associated with a greater risk of morbidity and mortality. Furthermore, decreased muscle mass, abnormal muscle function, and endothelial dysfunction in this patient population is known to contribute to disease progression. In adult patients with 2 ventricles and heart failure, reduced exercise capacity, muscle mass, and muscle strength are powerful predictors of poor outcomes, and exercise interventions can not only improve exercise capacity and muscle mass, but also reverse endothelial dysfunction. Despite these known benefits of exercise, pediatric Fontan patients do not exercise routinely due to their chronic condition, perceived restrictions to exercise, and parental overprotection. Limited exercise interventions in children with congenital heart disease have demonstrated that exercise is safe and effective; however, these studies have been conducted in small, heterogeneous groups, and most had few Fontan patients. Critically, adherence is a major limitation in pediatric exercise interventions delivered on-site, with adherence rates as low as 10%, due to distance from site, transportation difficulties, and missed school or workdays. To overcome these challenges, we utilize live-video conferencing to deliver the supervised exercise sessions. Our multidisciplinary team of experts will assess the effectiveness of a live-video-supervised exercise intervention, rigorously designed to maximize adherence, and improve key and novel measures of health in pediatric Fontan patients associated with poor long-term outcomes. Our ultimate goal is the translation of this model to clinical application as an "exercise prescription" to intervene early in pediatric Fontan patients and decrease long-term morbidity and mortality.

Accelerating pandemic response with the emergency Omicron RT-PCR test: A comprehensive solution for COVID-19 diagnosis and tracking.

The Omicron variant of concern (VOC) has surged in many countries and replaced the previously reported VOC. To identify different Omicron strains/sublineages on a rapid, convenient, and precise platform, we report a novel multiplex real-time reverse transcriptase polymerase chain reaction (RT-PCR) method in one tube based on the Omicron lineage sequence variants' information. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) subvariants were used in a PCR-based assay for rapid identification of Omicron sublineage genotyping in 1000 clinical samples. Several characteristic mutations were analyzed using specific primers and probes for the spike gene, del69-70, and F486V. To distinguish Omicron sublineages (BA.2, BA.4, and BA.5), the NSP1:141-143del in the ORF1a region and D3N mutation in membrane protein occurring outside the spike protein region were analyzed. Results from the real-time PCR assay for one-tube accuracy were compared to those of whole genome sequencing. The developed PCR assay was used to analyze 400 SARS-CoV-2 positive samples. Ten samples determined as BA.4 were positive for NSP1:141-143del, del69-70, and F486V mutations; 160 BA.5 samples were positive for D3N, del69-70, and F486V mutations, and 230 BA.2 samples were without del69-70. Screening these samples allowed the identification of epidemic trends at different time intervals. Our novel one-tube multiplex PCR assay was effective in identifying Omicron sublineages.

Point-of-care testing for adult out-of-hospital cardiac arrest resuscitated at the ED to predict ROSC: Development and external validation of POC-ED-ROSC model.

BACKGROUND AND IMPORTANCE: Most prediction models, like return of spontaneous circulation (ROSC) after cardiac arrest (RACA) or Utstein-based (UB)-ROSC score, were developed for prehospital settings to predict the probability of ROSC in patients with out-of-hospital cardiac arrest (OHCA). A prediction model has been lacking for the probability of ROSC in patients with OHCA at emergency departments (EDs). OBJECTIVE: In the present study, a point-of-care (POC) testing-based model, POC-ED-ROSC, was developed and validated for predicting ROSC of OHCA at EDs. DESIGN, SETTINGS AND PARTICIPANTS: Prospectively collected data for adult OHCA patients between 2015 and 2020 were analysed. POC blood gas analysis obtained within 5 min of ED arrival was used. OUTCOMES MEASURE AND ANALYSIS: The primary outcome was ROSC. In the derivation cohort, multivariable logistic regression was used to develop the POC-ED-ROSC model. In the temporally split validation cohort, the discriminative performance of the POC-ED-ROSC model was assessed using the area under the receiver operating characteristic (ROC) curve (AUC) and compared with RACA or UB-ROSC score using DeLong test. MAIN RESULTS: The study included 606 and 270 patients in the derivation and validation cohorts, respectively. In the total cohort, 471 patients achieved ROSC. Age, initial cardiac rhythm at ED, pre-hospital resuscitation duration, and POC testing-measured blood levels of lactate, potassium and glucose were significant predictors included in the POC-ED-ROSC model. The model was validated with fair discriminative performance (AUC: 0.75, 95% confidence interval [CI]: 0.69-0.81) with no significant differences from RACA (AUC: 0.68, 95% CI: 0.62-0.74) or UB-ROSC score (AUC: 0.74, 95% CI: 0.68-0.79). CONCLUSION: Using only six easily accessible variables, the POC-ED-ROSC model can predict ROSC for OHCA resuscitated at ED with fair accuracy.

Acute exacerbation of chronic obstructive pulmonary disease in United States emergency departments, 2010-2018.

OBJECTIVES: Little is known about the recent status of acute exacerbation of chronic obstructive pulmonary disease (AECOPD) in the U.S. emergency department (ED). This study aimed to describe the disease burden (visit and hospitalization rate) of AECOPD in the ED and to investigate factors associated with the disease burden of AECOPD. METHODS: Data were obtained from the National Hospital Ambulatory Medical Care Survey (NHAMCS), 2010-2018. Adult ED visits (aged 40 years or above) with AECOPD were identified using International Classification of Diseases codes. Analysis used descriptive statistics and multivariable logistic regression accounting for NHAMCS's complex survey design. RESULTS: There were 1,366 adult AECOPD ED visits in the unweighted sample. Over the 9-year study period, there were an estimated 7,508,000 ED visits for AECOPD, and the proportion of AECOPD visits in the entire ED population remained stable at approximately 14 per 1,000 visits. The mean age of these AECOPD visits was 66 years, and 42% were men. Medicare or Medicaid insurance, presentation in non-summer seasons, the Midwest and South regions (vs. Northeast), and arrival by ambulance were independently associated with a higher visit rate of AECOPD, whereas non-Hispanic black or Hispanic race/ethnicity (vs. non-Hispanic white) was associated with a lower visit rate of AECOPD. The proportion of AECOPD visits that were hospitalized decreased from 51% to 2010 to 31% in 2018 (p = 0.002). Arrival by ambulance was independently associated with a higher hospitalization rate, whereas the South and West regions (vs. Northeast) were independently associated with a lower hospitalization rate. The use of antibiotics appeared to be stable over time, but the use of systemic corticosteroids appeared to increase with near statistical significance (p = 0.07). CONCLUSIONS: The number of ED visits for AECOPD remained high; however, hospitalizations for AECOPD appeared to decrease over time. Some patients were disproportionately affected by AECOPD, and certain patient and ED factors were associated with hospitalizations. The reasons for decreased ED admissions for AECOPD deserve further investigation.

An AI-Enabled Dynamic Risk Stratification for Emergency Department Patients with ECG and CXR Integration.

Emergency department (ED) triage scale determines the priority of patient care and foretells the prognosis. However, the information retrieved from the initial assessment is limited, hindering the risk identification accuracy of triage. Therefore, we sought to develop a 'dynamic' triage system as secondary screening, using artificial intelligence (AI) techniques to integrate information from initial assessment data and subsequent examinations. This retrospective cohort study included 134,112 ED visits with at least one electrocardiography (ECG) and chest X-ray (CXR) in a medical center from 2012 to 2022. Additionally, an independent community hospital provided 45,614 ED visits as an external validation set. We trained an eXtreme gradient boosting (XGB) model using initial assessment data to predict all-cause mortality in 7 days. Two deep learning models (DLMs) using ECG and CXR were trained to stratify mortality risks. The dynamic triage levels were based on output from the XGB-triage and DLMs from ECG and CXR. During the internal and external validation, the area under the receiver operating characteristic curve (AUC) of the XGB-triage model was >0.866; furthermore, the AUCs of DLMs using ECG and CXR were >0.862 and >0.886, respectively. The dynamic triage scale provided a higher C-index (0.914-0.920 vs. 0.827-0.843) than the original one and demonstrated better predictive ability for 5-year mortality, 30-day ED revisit, and 30-day discharge. The AI-based risk scale provides a more accurate and dynamic stratification of mortality risk in ED patients, particularly in identifying patients who tend to be overlooked due to atypical symptoms.

Deep Learning-based Diagnosis and Localization of Pneumothorax on Portable Supine Chest X-ray in Intensive and Emergency Medicine: A Retrospective Study.

PURPOSE: To develop two deep learning-based systems for diagnosing and localizing pneumothorax on portable supine chest X-rays (SCXRs). METHODS: For this retrospective study, images meeting the following inclusion criteria were included: (1) patient age ≥ 20 years; (2) portable SCXR; (3) imaging obtained in the emergency department or intensive care unit. Included images were temporally split into training (1571 images, between January 2015 and December 2019) and testing (1071 images, between January 2020 to December 2020) datasets. All images were annotated using pixel-level labels. Object detection and image segmentation were adopted to develop separate systems. For the detection-based system, EfficientNet-B2, DneseNet-121, and Inception-v3 were the architecture for the classification model; Deformable DETR, TOOD, and VFNet were the architecture for the localization model. Both classification and localization models of the segmentation-based system shared the UNet architecture. RESULTS: In diagnosing pneumothorax, performance was excellent for both detection-based (Area under receiver operating characteristics curve [AUC]: 0.940, 95% confidence interval [CI]: 0.907-0.967) and segmentation-based (AUC: 0.979, 95% CI: 0.963-0.991) systems. For images with both predicted and ground-truth pneumothorax, lesion localization was highly accurate (detection-based Dice coefficient: 0.758, 95% CI: 0.707-0.806; segmentation-based Dice coefficient: 0.681, 95% CI: 0.642-0.721). The performance of the two deep learning-based systems declined as pneumothorax size diminished. Nonetheless, both systems were similar or better than human readers in diagnosis or localization performance across all sizes of pneumothorax. CONCLUSIONS: Both deep learning-based systems excelled when tested in a temporally different dataset with differing patient or image characteristics, showing favourable potential for external generalizability.

Isolation and Quantification of Methylated Cell-Free DNA in Plasma on an Integrated Microfluidic System.

Methylated cell-free DNA (cfDNA) has been deemed a promising biomarker for ovarian cancer (OvCa) prognosis and therapy selection. However, exploring the methylation profiles of tumor suppressor genes in cfDNA remains a challenge due to their extremely low concentrations and complicated protocols, as well as methodological constraints. In this study, an integrated microfluidic system was developed to automatically (1) capture methylated cfDNA in plasma by magnetic beads coated with the methyl-CpG-binding domain and (2) quantify the methylation level of tumor suppressor genes by on-chip quantitative polymerase chain reaction (qPCR). For capturing methylated cfDNA from a very small amount of plasma, samples along with beads were mixed in a new micromixer to enhance the capture rate. With a high capture rate (72%) and a limit of quantification of 0.1 pg/µL (3 orders of magnitude lower than that of the benchtop method), the compact system could detect the methylated cfDNA from only 20 µL of plasma sample in 2 h. Furthermore, the dynamic range, from 0.1 to 2000 pg/µL of methylated cfDNA, spans the physiological range in plasma, signifying that this device has great potential for personalized medicine in OvCa.

Integrated Microfluidic System for Cell-Free DNA Extraction from Plasma for Mutant Gene Detection and Quantification.

Ovarian cancer (OvCa) is among the most severe gynecologic cancers, yet individuals may be asymptomatic during its early stages. Routine, early screening for genetic abnormalities associated with OvCa could improve prognoses, and this can be achieved by detecting mutant genes in cell-free DNA (cfDNA). Herein, we developed an integrated microfluidic chip (IMC) that could extract cfDNA from plasma and automatically detect and quantify mutations in the OvCa biomarker BRCA1. The cfDNA extraction module relied on a vortex-type micromixer to mix cfDNA with magnetic beads surface-coated with cfDNA probes and could isolate 76% of molecules from a 200 µL plasma sample in 45 min. The cfDNA quantification module, which comprised a micropump that evenly distributed 4.5 µL of purified cfDNA into the on-chip, allele-specific quantitative polymerase chain reaction (qPCR) zones, was capable of quantifying mutant genes within 90 min. By automating the cfDNA extraction and qPCR processes, this IMC could be used for clinical screening for OvCa-associated mutations.

Genome-wide association study reveals ethnicity-specific SNPs associated with ankylosing spondylitis in the Taiwanese population.

BACKGROUND: Ankylosing spondylitis (AS) is an autoimmune disease affecting mainly spine and sacroiliac joints and adjacent soft tissues. Genome-wide association studies (GWASs) are used to evaluate genetic associations and to predict genetic risk factors that determine the biological basis of disease susceptibility. We aimed to explore the race-specific SNP susceptibility of AS in Taiwanese individuals and to investigate the association between HLA-B27 and AS susceptibility SNPs in Taiwan. METHODS: Genotyping data were collected from a medical center participating in the Taiwan Precision Medicine Initiative (TPMI) in the northern district of Taiwan. We designed a case-control study to identify AS susceptibility SNPs through GWAS. We searched the genome browser to find the corresponding susceptibility genes and used the GTEx database to confirm the regulation of gene expression. A polygenic risk score approach was also applied to evaluate the genetic variants in the prediction of developing AS. RESULTS: The results showed that the SNPs located on the sixth chromosome were related to higher susceptibility in the AS group. There was no overlap between our results and the susceptibility SNPs found in other races. The 12 tag SNPs located in the MHC region that were found through the linkage disequilibrium method had higher gene expression. Furthermore, Taiwanese people with HLA-B27 positivity had a higher proportion of minor alleles. This might be the reason that the AS prevalence is higher in Taiwan than in other countries. We developed AS polygenic risk score models with six different methods in which those with the top 10% polygenic risk had a fivefold increased risk of developing AS compared to the remaining group with low risk. CONCLUSION: A total of 147 SNPs in the Taiwanese population were found to be statistically significantly associated with AS on the sixth pair of chromosomes and did not overlap with previously published sites in the GWAS Catalog. Whether those genes mapped by AS-associated SNPs are involved in AS and what the pathogenic mechanism of the mapped genes is remain to be further studied.

Curbing COVID-19 progression and mortality with traditional Chinese medicine among hospitalized patients with COVID-19: A propensity score-matched analysis.

BACKGROUND: Viral- and host-targeted traditional Chinese medicine (TCM) formulae NRICM101 and NRICM102 were administered to hospitalized patients with COVID-19 during the mid-2021 outbreak in Taiwan. We report the outcomes by measuring the risks of intubation or admission to intensive care unit (ICU) for patients requiring no oxygen support, and death for those requiring oxygen therapy. METHODS: This multicenter retrospective study retrieved data of 840 patients admitted to 9 hospitals between May 1 and July 26, 2021. After propensity score matching, 302 patients (151 received NRICM101 and 151 did not) and 246 patients (123 received NRICM102 and 123 did not) were included in the analysis to assess relative risks. RESULTS: During the 30-day observation period, no endpoint occurred in the patients receiving NRICM101 plus usual care while 14 (9.27%) in the group receiving only usual care were intubated or admitted to ICU. The numbers of deceased patients were 7 (5.69%) in the group receiving NRICM102 plus usual care and 27 (21.95%) in the usual care group. No patients receiving NRICM101 transitioned to a more severe status; NRICM102 users were 74.07% less likely to die than non-users (relative risk= 25.93%, 95% confidence interval 11.73%-57.29%). CONCLUSION: NRICM101 and NRICM102 were significantly associated with a lower risk of intubation/ICU admission or death among patients with mild-to-severe COVID-19. This study provides real-world evidence of adopting broad-spectrum oral therapeutics and shortening the gap between outbreak and effective response. It offers a new vision in our preparation for future pandemics.