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.

Computer-Aided Diagnosis of Duchenne Muscular Dystrophy Based on Texture Pattern Recognition on Ultrasound Images Using Unsupervised Clustering Algorithms and Deep Learning.

OBJECTIVE: The feasibility of using deep learning in ultrasound imaging to predict the ambulatory status of patients with Duchenne muscular dystrophy (DMD) was previously explored for the first time. The present study further used clustering algorithms for the texture reconstruction of ultrasound images of DMD data sets and analyzed the difference in echo intensity between disease stages. METHODS: k-means (Kms) and fuzzy c-means (FCM) clustering algorithms were used to reconstruct the DMD data-set textures. Each image was reconstructed using seven texture-feature categories, six of which were used as the primary analysis items. The task of automatically identifying the ambulatory function and DMD severity was performed by establishing a machine-learning model. RESULTS: The experimental results indicated that the Gaussian Naïve Bayes and k-nearest neighbors classification models achieved an accuracy of 86.78% in ambulatory function classification. The decision-tree model achieved an identification accuracy of 83.80% in severity classification. A deep convolutional neural network model was established as the main structure of the deep-learning model while automatic auxiliary interpretation tasks of ambulatory function and severity were performed, and data augmentation was used to improve the recognition performance of the trained model. Both the visual geometry group (VGG)-16 and VGG-19 models achieved 98.53% accuracy in ambulatory-function classification. The VGG-19 model achieved 92.64% accuracy in severity classification. CONCLUSION: Regarding the overall results, the Kms and FCM clustering algorithms were used in this study to reconstruct the characteristic texture of the gastrocnemius muscle group in DMD, which was indeed helpful in quantitatively analyzing the deterioration of the gastrocnemius muscle group in patients with DMD at different stages. Subsequent combination of machine-learning and deep-learning technologies can automatically and accurately assist in identifying DMD symptoms and tracking DMD deterioration for long-term observation.

Prognostic value of arterial carbon dioxide tension during cardiopulmonary resuscitation in out-of-hospital cardiac arrest patients receiving extracorporeal resuscitation.

BACKGROUND: Current guidelines on extracorporeal cardiopulmonary resuscitation (ECPR) recommend careful patient selection, but precise criteria are lacking. Arterial carbon dioxide tension (PaCO2) has prognostic value in out-of-hospital cardiac arrest (OHCA) patients but has been less studied in patients receiving ECPR. We studied the relationship between PaCO2 during cardiopulmonary resuscitation (CPR) and neurological outcomes of OHCA patients receiving ECPR and tested whether PaCO2 could help ECPR selection. METHODS: This single-centre retrospective study enrolled 152 OHCA patients who received ECPR between January 2012 and December 2020. Favorable neurological outcome (FO) at discharge was the primary outcome. We used multivariable logistic regression to determine the independent variables for FO and generalised additive model (GAM) to determine the relationship between PaCO2 and FO. Subgroup analyses were performed to test discriminative ability of PaCO2 in subgroups of OHCA patients. RESULTS: Multivariable logistic regression showed that PaCO2 was independently associated with FO after adjusting for other favorable resuscitation characteristics (Odds ratio [OR] 0.23, 95% Confidence Interval [CI] 0.08-0.66, p-value = 0.006). GAM showed a near-linear reverse relationship between PaCO2 and FO. PaCO2 < 70 mmHg was the cutoff point for predicting FO. PaCO2 also had prognostic value in patients with less favorable characteristics, including non-shockable rhythm (OR, 3.78) or low flow time > 60 min (OR, 4.66). CONCLUSION: PaCO2 before ECMO implementation had prognostic value for neurological outcomes in OHCA patients. Patients with PaCO2 < 70 mmHg had higher possibility of FO, even in those with non-shockable rhythm or longer low-flow duration. PaCO2 could serve as an ECPR selection criterion.

Nanoscale sorting of extracellular vesicles via optically-induced dielectrophoresis on an integrated microfluidic system.

We reported a microfluidic system for sorting of extracellular vesicles (EVs), which can house DNAs, RNAs, lipids, proteins, and metabolites that are important in intercellular communication. Their presence within bodily fluids has demonstrated potential in both clinical diagnostic and therapeutic applications. Furthermore, EVs exhibit distinct subtypes categorized by their sizes, each endowed with unique biophysical properties. Despite several existing techniques for EV isolation and purification, diminished purity and prolonged processing times still hamper clinical utility; comprehensive capture of EVs remains an ongoing pursuit. To address these challenges, we devised an innovative method for automated sorting of nano-scale EVs employing optically-induced dielectrophoresis on an integrated microfluidic chip. With this approach, EVs of three distinct size categories (small: 100-150 nm, medium-sized: 150-225 nm, and large: 225-350 nm) could be isolated at a purity of 86%. This new method has substantial potential in expediting EV research and diagnostics.

Prognostic implication of heart failure stage and left ventricular ejection fraction for patients with in-hospital cardiac arrest: a 16-year retrospective cohort study.

BACKGROUND: The 2022 AHA/ACC/HFSA guidelines for the management of heart failure (HF) makes therapeutic recommendations based on HF status. We investigated whether the prognosis of in-hospital cardiac arrest (IHCA) could be stratified by HF stage and left ventricular ejection fraction (LVEF). METHODS: This single-center retrospective study analyzed the data of patients who experienced IHCA between 2005 and 2020. Based on admission diagnosis, past medical records, and pre-arrest echocardiography, patients were classified into general IHCA, at-risk for HF, pre-HF, HF with preserved ejection fraction (HFpEF), and HF with mildly reduced ejection fraction or HF with reduced ejection fraction (HFmrEF-or-HFrEF) groups. RESULTS: This study included 2,466 patients, including 485 (19.7%), 546 (22.1%), 863 (35.0%), 342 (13.9%), and 230 (9.3%) patients with general IHCA, at-risk for HF, pre-HF, HFpEF, and HFmrEF-or-HFrEF, respectively. A total of 405 (16.4%) patients survived to hospital discharge, with 228 (9.2%) patients achieving favorable neurological recovery. Multivariable logistic regression analysis indicated that pre-HF and HFpEF were associated with better neurological (pre-HF, OR: 2.11, 95% confidence interval [CI]: 1.23-3.61, p = 0.006; HFpEF, OR: 1.90, 95% CI: 1.00-3.61, p = 0.05) and survival outcomes (pre-HF, OR: 2.00, 95% CI: 1.34-2.97, p < 0.001; HFpEF, OR: 1.91, 95% CI: 1.20-3.05, p = 0.007), compared with general IHCA. CONCLUSION: HF stage and LVEF could stratify patients with IHCA into different prognoses. Pre-HF and HFpEF were significantly associated with favorable neurological and survival outcomes after IHCA. Further studies are warranted to investigate whether HF status-directed management could improve IHCA outcomes.

Application of Machine Learning to Ultrasonography in Identifying Anatomical Landmarks for Cricothyroidotomy Among Female Adults: A Multi-center Prospective Observational Study.

We aimed to develop machine learning (ML)-based algorithms to assist physicians in ultrasound-guided localization of cricoid cartilage (CC) and thyroid cartilage (TC) in cricothyroidotomy. Adult female volunteers were prospectively recruited from two hospitals between September and December, 2020. Ultrasonographic images were collected via a modified longitudinal technique. You Only Look Once (YOLOv5s), Faster Regions with Convolutional Neural Network features (Faster R-CNN), and Single Shot Detector (SSD) were selected as the model architectures. A total of 488 women (mean age: 36.0 years) participated in the study, contributing to a total of 292,053 frames of ultrasonographic images. The derived ML-based algorithms demonstrated excellent discriminative performance for the presence of CC (area under the receiver operating characteristic curve [AUC]: YOLOv5s, 0.989, 95% confidence interval [CI]: 0.982-0.994; Faster R-CNN, 0.986, 95% CI: 0.980-0.991; SSD, 0.968, 95% CI: 0.956-0.977) and TC (AUC: YOLOv5s, 0.989, 95% CI: 0.977-0.997; Faster R-CNN, 0.981, 95% CI: 0.965-0.991; SSD, 0.982, 95% CI: 0.973-0.990). Furthermore, in the frames where the model could correctly indicate the presence of CC or TC, it also accurately localized CC (intersection-over-union: YOLOv5s, 0.753, 95% CI: 0.739-0.765; Faster R-CNN, 0.720, 95% CI: 0.709-0.732; SSD, 0.739, 95% CI: 0.726-0.751) or TC (intersection-over-union: YOLOv5s, 0.739, 95% CI: 0.722-0.755; Faster R-CNN, 0.709, 95% CI: 0.687-0.730; SSD, 0.713, 95% CI: 0.695-0.730). The ML-based algorithms could identify anatomical landmarks for cricothyroidotomy in adult females with favorable discriminative and localization performance. Further studies are warranted to transfer this algorithm to hand-held portable ultrasound devices for clinical use.

Deep Learning-based Diagnosis of Pulmonary Tuberculosis on Chest X-ray in the Emergency Department: A Retrospective Study.

Prompt and correct detection of pulmonary tuberculosis (PTB) is critical in preventing its spread. We aimed to develop a deep learning-based algorithm for detecting PTB on chest X-ray (CXRs) in the emergency department. This retrospective study included 3498 CXRs acquired from the National Taiwan University Hospital (NTUH). The images were chronologically split into a training dataset, NTUH-1519 (images acquired during the years 2015 to 2019; n = 2144), and a testing dataset, NTUH-20 (images acquired during the year 2020; n = 1354). Public databases, including the NIH ChestX-ray14 dataset (model training; 112,120 images), Montgomery County (model testing; 138 images), and Shenzhen (model testing; 662 images), were also used in model development. EfficientNetV2 was the basic architecture of the algorithm. Images from ChestX-ray14 were employed for pseudo-labelling to perform semi-supervised learning. The algorithm demonstrated excellent performance in detecting PTB (area under the receiver operating characteristic curve [AUC] 0.878, 95% confidence interval [CI] 0.854-0.900) in NTUH-20. The algorithm showed significantly better performance in posterior-anterior (PA) CXR (AUC 0.940, 95% CI 0.912-0.965, p-value < 0.001) compared with anterior-posterior (AUC 0.782, 95% CI 0.644-0.897) or portable anterior-posterior (AUC 0.869, 95% CI 0.814-0.918) CXR. The algorithm accurately detected cases of bacteriologically confirmed PTB (AUC 0.854, 95% CI 0.823-0.883). Finally, the algorithm tested favourably in Montgomery County (AUC 0.838, 95% CI 0.765-0.904) and Shenzhen (AUC 0.806, 95% CI 0.771-0.839). A deep learning-based algorithm could detect PTB on CXR with excellent performance, which may help shorten the interval between detection and airborne isolation for patients with PTB.

Host-Pathogen Interactions in K. pneumoniae Urinary Tract Infections: Investigating Genetic Risk Factors in the Taiwanese Population.

BACKGROUND: Klebsiella pneumoniae (K. pneumoniae) urinary tract infections pose a significant challenge in Taiwan. The significance of this issue arises because of the growing concerns about the antibiotic resistance of K. pneumoniae. Therefore, this study aimed to uncover potential genomic risk factors in Taiwanese patients with K. pneumoniae urinary tract infections through genome-wide association studies (GWAS). METHODS: Genotyping data are obtained from participants with a history of urinary tract infections enrolled at the Tri-Service General Hospital as part of the Taiwan Precision Medicine Initiative (TPMI). A case-control study employing GWAS is designed to detect potential susceptibility single-nucleotide polymorphisms (SNPs) in patients with K. pneumoniae-related urinary tract infections. The associated genes are determined using a genome browser, and their expression profiles are validated via the GTEx database. The GO, Reactome, DisGeNET, and MalaCards databases are also consulted to determine further connections between biological functions, molecular pathways, and associated diseases between these genes. RESULTS: The results identified 11 genetic variants with higher odds ratios compared to controls. These variants are implicated in processes such as adhesion, protein depolymerization, Ca2+-activated potassium channels, SUMOylation, and protein ubiquitination, which could potentially influence the host immune response. CONCLUSIONS: This study implies that certain risk variants may be linked to K. pneumoniae infections by affecting diverse molecular functions that can potentially impact host immunity. Additional research and follow-up studies are necessary to elucidate the influence of these risk variants on infectious diseases and develop targeted interventions for mitigating the spread of K. pneumoniae urinary tract infections.

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.

Development and External Validation of Clinical Features-based Machine Learning Models for Predicting COVID-19 in the Emergency Department.

Introduction: Timely diagnosis of patients affected by an emerging infectious disease plays a crucial role in treating patients and avoiding disease spread. In prior research, we developed an approach by using machine learning (ML) algorithms to predict serious acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection based on clinical features of patients visiting an emergency department (ED) during the early coronavirus 2019 (COVID-19) pandemic. In this study, we aimed to externally validate this approach within a distinct ED population. Methods: To create our training/validation cohort (model development) we collected data retrospectively from suspected COVID-19 patients at a US ED from February 23-May 12, 2020. Another dataset was collected as an external validation (testing) cohort from an ED in another country from May 12-June 15, 2021. Clinical features including patient demographics and triage information were used to train and test the models. The primary outcome was the confirmed diagnosis of COVID-19, defined as a positive reverse transcription polymerase chain reaction test result for SARS-CoV-2. We employed three different ML algorithms, including gradient boosting, random forest, and extra trees classifiers, to construct the predictive model. The predictive performances were evaluated with the area under the receiver operating characteristic curve (AUC) in the testing cohort. Results: In total, 580 and 946 ED patients were included in the training and testing cohorts, respectively. Of them, 98 (16.9%) and 180 (19.0%) were diagnosed with COVID-19. All the constructed ML models showed acceptable discrimination, as indicated by the AUC. Among them, random forest (0.785, 95% confidence interval [CI] 0.747-0.822) performed better than gradient boosting (0.774, 95% CI 0.739-0.811) and extra trees classifier (0.72, 95% CI 0.677-0.762). There was no significant difference between the constructed models. Conclusion: Our study validates the use of ML for predicting COVID-19 in the ED and demonstrates its potential for predicting emerging infectious diseases based on models built by clinical features with temporal and spatial heterogeneity. This approach holds promise for scenarios where effective diagnostic tools for an emerging infectious disease may be lacking in the future.

A self-driven, microfluidic, integrated-circuit biosensing chip for detecting four cardiovascular disease biomarkers.

Cardiovascular diseases (CVDs) claimed the lives of nearly 21 million people worldwide in 2021, accounting for 30% of global deaths. However, one in five CVD patients is unaware that they have the disease, emphasizing the need for accurate biomarker monitoring. Herein we developed an integrated microfluidic system (IMS) for rapid quantification of four CVD biomarkers, including N-terminal pro B-type natriuretic peptide (NT-proBNP), fibrinogen, cardiac troponin I (cTnI), and C-reactive protein (CRP)- via aptamer-coated interdigitated electrodes (IDE) with integrated circuits (IC) and a self-driven IMS for sample treatment. The device was composed of plasma filtration, metering, and fluidic delay modules, and the former could extract 45% of plasma from a 20-µL blood sample; the metering module could quantify 5 µL of plasma within 90 s. Subsequently, the plasma was transported to a detection chamber, where IC-based IDE sensors made measurements within 5 min. The entire 15-min process allowed us to evaluate biomarkers across a wide dynamic range: NT-proBNP (0.1-10,000 pg/mL), fibrinogen (50-1,000 mg/dL), cTnI (0.1-10,000 pg/mL), and CRP (0.5-9 mg/L). Given that spiked blood samples were measured with reasonable accuracy (>80%), the IMS could see utility in CVD risk assessment and personalized medicine.

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.

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.

The incidence, predictors, and causes of cardiac arrest in United States emergency departments.

Background: Emergency department cardiac arrest (EDCA) is a global public health challenge associated with high mortality rates and poor neurological outcomes. This study aimed to describe the incidence, risk factors, and causes of EDCA during emergency department (ED) visits in the U.S. Methods: This retrospective cohort study used data from the 2019 Nationwide Emergency Department Sample (NEDS). Adult ED visits with EDCA were identified using the cardiopulmonary resuscitation code. We used descriptive statistics and multivariable logistic regression, considering NEDS's complex survey design. The primary outcome measure was EDCA incidence. Results: In 2019, there were approximately 232,000 ED visits with cardiac arrest in the U.S. The incidence rate of EDCA was approximately 0.2%. Older age, being male, black race, low median household income, weekend ED visits, having Medicare insurance, and ED visits in non-summer seasons were associated with a higher risk of EDCA. Hispanic race was associated with a lower risk of EDCA. Certain comorbidities (e.g., diabetes and cancer), trauma centers, hospitals with a metropolitan and/or teaching program, and hospitals in the South were associated with a higher risk of EDCA. Depression, dementia, and hypothyroidism were associated with a lower risk of EDCA. Septicemia, acute myocardial infarction, and respiratory failure, followed by drug overdose, were the predominant causes of EDCA. Conclusions: Some patients were disproportionately affected by EDCA. Strategies should be developed to target these modifiable risk factors, specifically factors within ED's control, to reduce the subsequent disease burden.

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.

Opportunistic Screening for Asymptomatic Left Ventricular Dysfunction With the Use of Electrocardiographic Artificial Intelligence: A Cost-Effectiveness Approach.

BACKGROUND: The burden of asymptomatic left ventricular dysfunction (LVD) is greater than that of heart failure; however, a cost-effective tool for asymptomatic LVD screening has not been well validated. We aimed to prospectively validate an artificial intelligence (AI)-enabled electrocardiography (ECG) algorithm for asymptomatic LVD detection and evaluate its cost-effectiveness for opportunistic screening. METHODS: In this prospective observational study, patients undergoing ECG at outpatient clinics or health check-ups were enrolled in 2 hospitals in Taiwan. Patients were stratified into LVD (left ventricular ejection fraction ≤ 40%) risk groups according to a previously developed ECG algorithm. The performance of AI-ECG was used to conduct a cost-effectiveness analysis of LVD screening compared with no screening. Incremental cost-effectiveness ratio (ICER) and sensitivity analyses were used to examine the cost-effectiveness and robustness of the results. RESULTS: Among the 29,137 patients, the algorithm demonstrated areas under the receiver operating characteristic curves of 0.984 and 0.945 for detecting LVD within 28 days in the 2 hospital cohorts. For patients not initially scheduled for ECG, the algorithm predicted future echocardiograms (high-risk, 46.2%; medium-risk, 31.4%; low-risk, 14.6%) and LVD (high-risk, 26.2%; medium-risk, 3.4%; low-risk, 0.1%) at 12 months. Opportunistic screening with AI-ECG could result in a negative ICER of -$7,439 for patients aged 65 years, with consistent cost-savings across age groups and particularly in men. Approximately 91.5% of the cases were found to be cost-effective at the willingness-to-pay threshold of $30,000 in the probabilistic analysis. CONCLUSIONS: The use of AI-ECG for asymptomatic LVD risk stratification is promising, and opportunistic screening in outpatient clinics has the potential to reduce costs.

GRAIL gene knockout mice protect against aging-related and noise-induced hearing loss.

BACKGROUND: Hearing loss is a global health issue and its etiopathologies involve complex molecular pathways. The ubiquitin-proteasome system has been reported to be associated with cochlear development and hearing loss. The gene related to anergy in lymphocytes ( GRAIL ), as an E3 ubiquitin ligase, has not, as yet, been examined in aging-related and noise-induced hearing loss mice models. METHODS: This study used wild-type (WT) and GRAIL knockout (KO) mice to examine cochlear hair cells and synaptic ribbons using immunofluorescence staining. The hearing in WT and KO mice was detected using auditory brainstem response. Gene expression patterns were compared using RNA-sequencing to identify potential targets during the pathogenesis of noise-induced hearing loss in WT and KO mice. RESULTS: At the 12-month follow-up, GRAIL KO mice had significantly less elevation in threshold level and immunofluorescence staining showed less loss of outer hair cells and synaptic ribbons in the hook region compared with GRAIL WT mice. At days 1, 14, and 28 after noise exposure, GRAIL KO mice had significantly less elevation in threshold level than WT mice. After noise exposure, GRAIL KO mice showed less loss of outer hair cells in the cochlear hook and basal regions compared with WT mice. Moreover, immunofluorescence staining showed less loss of synaptic ribbons in the hook regions of GRAIL KO mice than of WT mice. RNA-seq analysis results showed significant differences in C-C motif chemokine ligand 19 ( CCL19 ), C-C motif chemokine ligand 21 ( CCL21 ), interleukin 25 ( IL25 ), glutathione peroxidase 6 ( GPX6 ), and nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 1 ( NOX1 ) genes after noise exposure. CONCLUSION: The present data demonstrated that GRAIL deficiency protects against aging-related and noise-induced hearing loss. The mechanism involved needs to be further clarified from the potential association with synaptic modulation, inflammation, and oxidative stress.

Artificial Intelligence-Enabled Electrocardiography Detects B-Type Natriuretic Peptide and N-Terminal Pro-Brain Natriuretic Peptide.

BACKGROUND: The B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (pBNP) are predictors of cardiovascular morbidity and mortality. Since the artificial intelligence (AI)-enabled electrocardiogram (ECG) system is widely used in the management of many cardiovascular diseases (CVDs), patients requiring intensive monitoring may benefit from an AI-ECG with BNP/pBNP predictions. This study aimed to develop an AI-ECG to predict BNP/pBNP and compare their values for future mortality. METHODS: The development, tuning, internal validation, and external validation sets included 47,709, 16,249, 4001, and 6042 ECGs, respectively. Deep learning models (DLMs) were trained using a development set for estimating ECG-based BNP/pBNP (ECG-BNP/ECG-pBNP), and the tuning set was used to guide the training process. The ECGs in internal and external validation sets belonging to nonrepeating patients were used to validate the DLMs. We also followed-up all-cause mortality to explore the prognostic value. RESULTS: The DLMs accurately distinguished mild (≥500 pg/mL) and severe (≥1000 pg/mL) an abnormal BNP/pBNP with AUCs of ≥0.85 in the internal and external validation sets, which provided sensitivities of 68.0-85.0% and specificities of 77.9-86.2%. In continuous predictions, the Pearson correlation coefficient between ECG-BNP and ECG-pBNP was 0.93, and they were both associated with similar ECG features, such as the T wave axis and correct QT interval. ECG-pBNP provided a higher all-cause mortality predictive value than ECG-BNP. CONCLUSIONS: The AI-ECG can accurately estimate BNP/pBNP and may be useful for monitoring the risk of CVDs. Moreover, ECG-pBNP may be a better indicator to manage the risk of future mortality.

Combined use of microbubbles of various sizes and single-transducer dual-frequency ultrasound for safe and efficient inner ear drug delivery.

We have previously applied ultrasound (US) with microbubbles (MBs) to enhance inner ear drug delivery, with most experiments conducted using single-frequency, high-power density US, and multiple treatments. In the present study, the treatment efficacy was enhanced and safety concerns were addressed using a combination of low-power-density, single-transducer, dual-frequency US (I SPTA = 213 mW/cm2) and MBs of different sizes coated with insulin-like growth factor 1 (IGF-1). This study is the first to investigate the drug-coating capacity of human serum albumin (HSA) MBs of different particle sizes and their drug delivery efficiency. The concentration of HSA was adjusted to produce different MB sizes. The drug-coating efficiency was significantly higher for large-sized MBs than for smaller MBs. In vitro Franz diffusion experiments showed that the combination of dual-frequency US and large MB size delivered the most IGF-1 (24.3 ± 0.47 ng/cm2) to the receptor side at the second hour of treatment. In an in vivo guinea pig experiment, the efficiency of IGF-1 delivery into the inner ear was 15.9 times greater in animals treated with the combination of dual-frequency US and large MBs (D-USMB) than in control animals treated with round window soaking (RWS). The IGF-1 delivery efficiency was 10.15 times greater with the combination of single-frequency US and large size MBs (S-USMB) than with RWS. Confocal microscopy of the cochlea showed a stronger distribution of IGF-1 in the basal turn in the D-USMB and S-USMB groups than in the RWS group. In the second and third turns, the D-USMB group showed the greatest IGF-1 distribution. Hearing assessments revealed no significant differences among the D-USMB, S-USMB, and RWS groups. In conclusion, the combination of single-transducer dual-frequency US and suitably sized MBs can significantly reduce US power density while enhancing the delivery of large molecular weight drugs, such as IGF-1, to the inner ear.

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.