Search

The intersection of COVID-19 and pulmonary embolism (PE) has posed unprecedented challenges in medical diagnostics. The critical nature of PE and its increased incidence during the pandemic underline the need for improved detection methods. This study evaluates the effectiveness of advanced deep learning techniques in enhancing PE detection in post-COVID-19 patients through Computed Tomography Pulmonary Angiography (CTPA) scans. Using a dataset of 746 anonymized CTPA images from 25 patients, we fine-tuned the state-of-the-art Ultralytics YOLOv8 object detection model, which was trained on 676 images with 1,517 annotated bounding boxes and validated on 70 images with 108 bounding boxes. After 200 epochs of training, which lasted approximately 1.021 hours, the YOLOv8 model demonstrated significant diagnostic proficiency, achieving a mean Average Precision (mAP) of 0.683 at an IoU threshold of 0.50 and a mAP of 0.246 at the IoU range of 0.50:0.95 in the validation dataset. Notably, the model reached a maximum precision of 0.85949 and a maximum recall of 0.81481, though these metrics were observed in separate epochs. These findings emphasize the model’s potential for high diagnostic accuracy and offer a promising direction for deploying AI tools in clinical settings, significantly contributing to healthcare innovation and patient care post-pandemic. [ABSTRACT FROM AUTHOR]

This study investigates the forecasting of cardiovascular mortality trends in Greece’s elderly population. Utilizing mortality data from 2001 to 2020, we employ two forecasting models: the Autoregressive Integrated Moving Average (ARIMA) and Facebook's Prophet model. Our study evaluates the efficacy of these models in predicting cardiovascular mortality trends over 2020-2030. The ARIMA model showcased predictive accuracy for the general and male population within the 65-79 age group, whereas the Prophet model provided better forecasts for females in the same age bracket. Our findings emphasize the need for adaptive forecasting tools that accommodate demographic-specific characteristics and highlight the role of advanced statistical methods in health policy planning. [ABSTRACT FROM AUTHOR]

In the realm of ophthalmic surgeries, silicone oil is often utilized as a tamponade agent for repairing retinal detachments, but it necessitates subsequent removal. This study harnesses the power of machine learning to analyze the macular and optic disc perfusion changes pre and post-silicone oil removal, using Optical Coherence Tomography Angiography (OCTA) data. Building upon the foundational work of prior research, our investigation employs Gaussian Process Regression (GPR) and Long Short-Term Memory (LSTM) networks to create predictive models based on OCTA scans. We conducted a comparative analysis focusing on the flow in the outer retina and vessel density in the deep capillary plexus (superior-hemi and perifovea) to track perfusion changes across different time points. Our findings indicate that while machine learning models predict the flow in the outer retina with reasonable accuracy, predicting the vessel density in the deep capillary plexus (particularly in the superior-hemi and perifovea regions) remains challenging. These results underscore the potential of machine learning to contribute to personalized patient care in ophthalmology, despite the inherent complexities in predicting ocular perfusion changes. [ABSTRACT FROM AUTHOR]

In an era increasingly focused on integrating Artificial Intelligence (AI) into healthcare, the utility and user satisfaction of AI applications like ChatGPT have become pivotal research areas. This study, conducted in Greece, engaged 193 doctors from various medical departments who interacted with ChatGPT 4.0 through a custom web application. The participants, representing a diverse range of medical specialties, received responses from the specific chatbot tailored to their specific departmental inquiries. Their satisfaction was gauged using a validated form featuring a 1-to-5 rating scale. The results highlighted a possible correlation between the doctors' medical departments and their satisfaction levels with ChatGPT 4.0. Significantly, doctors from certain departments (like General Surgery and Cardiology) reported lower satisfaction scores, ranging from 2.73 to 2.80 out of 5, in contrast to their colleagues from departments like Biopathology and Orthopedics, who scored between 4.00 and 4.46 out of 5. This variation in satisfaction levels underscores the diverse needs within different medical specialties and illuminates both the potential of ChatGPT and the areas needing improvement, especially in delivering department-specific medical information. Despite its limitations, ChatGPT version 4.0 is emerging as a valuable tool in the medical community, indicating potential future advancements and more extensive integration into healthcare practices. The study's findings are crucial in understanding the distinct preferences and requirements of healthcare professionals across various medical departments, thereby guiding the future development of AI tools in healthcare. [ABSTRACT FROM AUTHOR]

Simple Summary: Robot-assisted surgery facilitates the examination and improvement of artificial intelligence (AI) integration in surgical processes through the provision of comprehensive telemetry data and an advanced viewing interface. Machine learning (ML) techniques enhance the feedback on the development of surgical abilities, the efficacy of the surgical operation, surgical guiding, and predicted results. By incorporating tension sensors on the robotic arms and employing augmented reality techniques, the surgical experience can be greatly improved. This enables the continuous monitoring of organ movements in real time, resulting in enhanced precision and accuracy. The integration of artificial intelligence (AI) into robotic surgery is anticipated to have a substantial influence on the education of upcoming surgeons and improve the entire surgical process. Both endeavours strive for ultimate accuracy in order to enhance the quality of surgical care. With the rapid increase in computer processing capacity over the past two decades, machine learning techniques have been applied in many sectors of daily life. Machine learning in therapeutic settings is also gaining popularity. We analysed current studies on machine learning in robotic urologic surgery. We searched PubMed/Medline and Google Scholar up to December 2023. Search terms included "urologic surgery", "artificial intelligence", "machine learning", "neural network", "automation", and "robotic surgery". Automatic preoperative imaging, intraoperative anatomy matching, and bleeding prediction has been a major focus. Early artificial intelligence (AI) therapeutic outcomes are promising. Robot-assisted surgery provides precise telemetry data and a cutting-edge viewing console to analyse and improve AI integration in surgery. Machine learning enhances surgical skill feedback, procedure effectiveness, surgical guidance, and postoperative prediction. Tension-sensors on robotic arms and augmented reality can improve surgery. This provides real-time organ motion monitoring, improving precision and accuracy. As datasets develop and electronic health records are used more and more, these technologies will become more effective and useful. AI in robotic surgery is intended to improve surgical training and experience. Both seek precision to improve surgical care. AI in ''master–slave" robotic surgery offers the detailed, step-by-step examination of autonomous robotic treatments. [ABSTRACT FROM AUTHOR]

Purpose: The high incidence of urinary tract infections (UTIs), often in nosocomial environments, is a major cause of antimicrobial resistance (AMR). The dissemination of antibiotic-resistant infections results in very high health and economic burdens for patients and healthcare systems, respectively. This study aims to determine and present the antibiotic resistance profiles of the most common pathogens in a urology department in Greece. Methods: During the period 2019–2020, we included 12,215 clinical samples of blood and urine specimens that tested positive for the following pathogens: Escherichia coli, Enterococcus faecium, Enterococcus faecalis, Proteus mirabilis, Klebsiella pneumoniae, or Pseudomonas aeruginosa, as these are the most commonly encountered microbes in a urology department. Results: The analysis revealed a 22.30% mean resistance rate of E. coli strains with a 76.42% resistance to ampicillin and a 54.76% resistance rate to ciprofloxacin in the two-year period. It also showed an approximately 19% resistance rate of P. mirabilis strains and a mean resistance rate of 46.205% of K. pneumoniae strains, with a decreasing trend during the four semesters (p-value < 0.001), which presented an 80% resistance rate to ampicillin/sulbactam and 73.33% to ciprofloxacin. The resistance to carbapenems was reported to be 39.82%. The analysis revealed a 24.17% mean resistance rate of P. aeruginosa with a declining rate over the two-year period (p-value < 0.001). The P. aeruginosa strains were 38% resistant to fluoroquinolones and presented varying resistance against carbapenems (31.58% against doripenem and 19.79% against meropenem). Regarding the Enteroccocal strains, a 46.91% mean resistance was noted for E. faecium with 100% resistance to ampicillin, and a 24.247% mean resistance rate for E. faecalis strains that were 41% resistant to ciprofloxacin. Both types showed 100% sensitivity to linezolid. Conclusions: The dissemination of antibiotic-resistant pathogens poses the need to implement surveillance programs and, consequently, to develop strategies to prevent the emergence of such pathogens in order to optimize patient outcomes. [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) has shown the ability to enhance the accuracy and efficiency of physicians. ChatGPT is an AI chatbot that can interact with humans through text, over the internet. It is trained with machine learning algorithms, using large datasets. In this study, we compare the performance of using a ChatGPT API 3.5 Turbo model to a general model, in assisting urologists in obtaining accurate, valid medical information. The API was accessed through a Python script that was applied specifically for this study based on 2023 EAU guidelines in PDF format. This custom-trained model leads to providing doctors with more precise, prompt answers about specific urologic subjects, thus helping them, ultimately, providing better patient care. [ABSTRACT FROM AUTHOR]

ASCAPE Project is a study aiming to implement the advances of Artificial Intelligence (AI), to support prostate cancer survivors, regarding quality of life issues. The aim of the study is to determine characteristics of patients who accepted to join ASCAPE project. It results that participants of the study mainly originate from higher-educated societies that are better informed about the potential benefits of AI in medicine. Therefore, efforts should be focused on eliminating patients’ reluctancy by better informing them on the potential benefits of AI. [ABSTRACT FROM AUTHOR]

In this study a deep learning architecture based on a convolutional neural network has been evaluated for the classification of white light images of colorectal polyps acquired during the process of a colonoscopy, to estimate the accuracy of the optical recognition of histologic types of polyps. Convolutional neural networks (CNNs), a subclass of artificial neural networks that have gained dominance in several computer vision tasks, are gaining popularity in many medical fields, including endoscopy. The TensorFlow framework was used for implementing EfficientNetB7, which was trained with 924 images, drawn from 86 patients. 55% of the polyps were adenomas, 22% were hyperplastic, and 17% were lesions with sessile serrations. The validation loss, accuracy, and AUC ROC were 0.4845, 0.7778, and 0.8881 respectively. [ABSTRACT FROM AUTHOR]

The COVID-19 infection is still a serious threat to public health and healthcare systems. Numerous practical machine learning applications have been investigated in this context to support clinical decision-making, forecast disease severity and admission to the intensive care unit, as well as to predict the demand for hospital beds, equipment, and staff in the future. We retrospectively analyzed demographics, and routine blood biomarkers from consecutive Covid-19 patients admitted to the intensive care unit (ICU) of a public tertiary hospital, during a 17- month period, relative to the outcome, in order to build a prognostic model. We used the Google Vertex AI platform, on the one hand, to evaluate its performance in predicting ICU mortality, and on the other hand to show the ease with which even non-experts can make prognostic models. The model’s performance regarding the area under the receiver operating characteristic curve (AUC-ROC) was 0.955. The six highest-ranked predictors of mortality in the prognostic model were age, serum urea, platelets, C-reactive protein, hemoglobin, and SGOT. [ABSTRACT FROM AUTHOR]

In this AI-focused era, researchers are delving into AI applications in healthcare, with ChatGPT being a primary focus. This Greek study involved 182 doctors from various regions, utilizing a custom web application connected to ChatGPT 4.0. Doctors from diverse departments and experience levels engaged with ChatGPT, which provided tailored responses. Over a month, data was collected using a form with a 1-to-5 rating scale. The results showed varying satisfaction levels across four criteria: clarity, response time, accuracy, and overall satisfaction. ChatGPT's response speed received high ratings (3.85/5.0), whereas clarity of information was moderately rated (3.43/5.0). A significant observation was the correlation between a doctor's experience and their satisfaction with ChatGPT. More experienced doctors (over 21 years) reported lower satisfaction (2.80–3.74/5.0) compared to their less experienced counterparts (3.43–4.20/5.0). At the medical field level, Internal Medicine showed higher satisfaction in evaluation criteria (ranging from 3.56 to 3.88), compared to other fields, while Psychiatry scored higher overall, with ratings from 3.63 to 5.00. The study also compared two departments: Urology and Internal Medicine, with the latter being more satisfied with the accuracy, and clarity of provided information, response time, and overall compared to Urology. These findings illuminate the specific needs of the health sector and highlight both the potential and areas for improvement in ChatGPT's provision of specialized medical information. Despite current limitations, ChatGPT, in its present version, offers a valuable resource to the medical community, signaling further advancements and potential integration into healthcare practices. • Greek doctors evaluate ChatGPT 4.0 nationwide via a custom web app. • ChatGPT earns high marks for speed; clarity scores are moderate. • More experienced doctors report lower satisfaction. • Doctors in Internal Medicine and Psychiatry show higher satisfaction than those in Surgery or Special Treatment Units. • Findings underscore ChatGPT's potential and improvement areas in healthcare. [ABSTRACT FROM AUTHOR]

This study explores the relationship between psychological factors and children's BMI, using clustering methods like Gaussian Mixture Models and Spectral Clustering. Affinity Propagation was particularly effective, suggesting that tailored interventions based on psychological assessments could improve obesity management in children. [ABSTRACT FROM AUTHOR]

Hospital-acquired infections, particularly in ICU, are becoming more frequent in recent years, with the most serious of them being Gram-negative bacterial infections. Among them, Acinetobacter baumannii, Klebsiella pneumoniae, and Pseudomonas aeruginosa are considered the most resistant bacteria encountered in ICU and other wards. Given the fact that about 24 hours are usually required to perform common antibiotic resistance tests after the bacteria identification, the use of machine learning techniques could be an additional decision support tool in selecting empirical antibiotic treatment based on the sample type, bacteria, and patient's basic characteristics. In this article, five machine learning (ML) models were evaluated to predict antimicrobial resistance of Acinetobacter baumannii, Klebsiella pneumoniae, and Pseudomonas aeruginosa. We suggest implementing ML techniques to forecast antibiotic resistance using data from the clinical microbiology laboratory, available in the Laboratory Information System (LIS). [ABSTRACT FROM AUTHOR]

Multi-drug-resistant (MDR) infections and their devastating consequences constitute a global problem and a constant threat to public health with immense costs for their treatment. Early identification of the pathogen and its antibiotic resistance profile is crucial for a favorable outcome. Given the fact that more than 24 hours are usually required to perform common antibiotic resistance tests after the sample collection, the implementation of machine learning methods could be of significant help in selecting empirical antibiotic treatment based only on the sample type, Gram stain, and patient's basic characteristics. In this paper, five machine learning (ML) algorithms have been tested to determine antibiotic susceptibility predictions using simple demographic data of the patients, as well as culture results and antibiotic susceptibility tests. Implementing ML algorithms to antimicrobial susceptibility data may offer insightful antibiotic susceptibility predictions to assist clinicians in decision-making regarding empirical treatment. [ABSTRACT FROM AUTHOR]

Data sharing has become an increasingly common process among health organizations, but any organization will most likely try to hide some sensitive patterns before sharing its data with others. Local Distortion Hiding (LDH), a recently proposed algorithm, has been evaluated only on the assumption of an opponent using the J48 (C4.5) classification algorithm. We now extend the basic approach, and we present a medical dataset hiding case study of a processed by LDH and attacked with the CART algorithm. [ABSTRACT FROM AUTHOR]

Coronavirus disease (COVID-19) constitutes an ongoing global health problem with significant morbidity and mortality. It usually presents characteristic findings on a chest CT scan, which may lead to early detection of the disease. A timely and accurate diagnosis of COVID-19 is the cornerstone for the prompt management of the patients. The aim of the present study was to evaluate the performance of an automated machine learning algorithm in the diagnosis of Covid-19 pneumonia using chest CT scans. Diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC), sensitivity, and positive predictive value. The method's average precision was 0.932. We suggest that auto-ML platforms help users with limited ML expertise train image recognition models by only uploading the examined dataset and performing some basic settings. Such methods could deliver significant potential benefits for patients in the future by allowing for earlier disease detection and care. [ABSTRACT FROM AUTHOR]

Introduction: The aim of this study was to evaluate alterations in corneal astigmatism, axial anterior corneal curvature, anterior chamber depth, and central corneal thickness (CCT) two months after the unilateral recession of lateral rectus muscle in children. Methods: This prospective study included 37 children with intermittent exotropia who would undergo unilateral lateral rectus muscle recession. All measurements were performed using Pentacam®. Comparisons were made between the operated and fellow unoperated eyes, pre- and post-operatively. The assessment was made for changes in the radius of axial curvature on major meridians at 3 and 3.5 mm from the optical corneal center in the mid-peripheral zone. Astigmatism changes of the anterior and posterior corneal surface were calculated using vector analysis software (astigMATIC®). The interaction between age or CCT and postoperative changes in anterior and posterior surface corneal astigmatism were examined with ANOVA model. Results: In the intervention group, changes in anterior and posterior corneal surface astigmatism were statistically significant, with a mean increase of 0.56Dx90 and 0.08Dx87, respectively. In the mid-peripheral corneal zone, an increase was observed in the radius of anterior corneal axial curvature, more evident temporal 3 and 3.5 mm from the corneal center on the horizontal meridian, with corresponding decrease superiorly and inferiorly at 3 and 3.5 mm from the corneal center on the vertical meridian. Discussion: The changes in total astigmatism of the operated eyes are mainly attributed to the anterior corneal surface. These changes are associated with flattening in the 180 meridian of the cornea, leading to a shift to “with-the-rule” astigmatism. [ABSTRACT FROM AUTHOR]

Data sharing among health organizations has become an increasingly common process, but any organization will most likely try to hide some sensitive patterns before it shares its data with others. This article focuses on the protection of sensitive patterns when we assume that decision trees will be the models to be induced. We apply a heuristic approach to hide any arbitrary rule from the derivation of a binary decision tree. The proposed hiding method is preferred over other heuristic solutions such as output disturbance or encryption methods that limit data usability, as the raw data itself can then more easily be offered for access by any third parties. [ABSTRACT FROM AUTHOR]

Optimal antibiotic use for the treatment of nosocomial infections plays a central role in the effort to control the rapidly increasing prevalence of multidrugresistant bacteria. Antibiotic selection should be based on accurate knowledge of local susceptibility rates. Traditional methods of resistance reporting, which are in routine use by microbiology laboratories could be enhanced by using statistically significant results. We present a method of reporting based on antibiotic susceptibility data analysis which offers an accurate tool that reduces clinician uncertainty and enables optimization of the antibiotic selection process. [ABSTRACT FROM AUTHOR]

Background/Objectives : Melanoma, an aggressive form of skin cancer, accounts for a significant proportion of skin-cancer-related deaths worldwide. Early and accurate differentiation between melanoma and benign melanocytic nevi is critical for improving survival rates but remains challenging because of diagnostic variability. Convolutional neural networks (CNNs) have shown promise in automating melanoma detection with accuracy comparable to expert dermatologists. This study evaluates and compares the performance of four CNN architectures-DenseNet121, ResNet50V2, NASNetMobile, and MobileNetV2-for the binary classification of dermoscopic images. Methods : A dataset of 8825 dermoscopic images from DermNet was standardized and divided into training (80%), validation (10%), and testing (10%) subsets. Image augmentation techniques were applied to enhance model generalizability. The CNN architectures were pre-trained on ImageNet and customized for binary classification. Models were trained using the Adam optimizer and evaluated based on accuracy, area under the receiver operating characteristic curve (AUC-ROC), inference time, and model size. The statistical significance of the differences was assessed using McNemar's test. Results : DenseNet121 achieved the highest accuracy (92.30%) and an AUC of 0.951, while ResNet50V2 recorded the highest AUC (0.957). MobileNetV2 combined efficiency with competitive performance, achieving a 92.19% accuracy, the smallest model size (9.89 MB), and the fastest inference time (23.46 ms). NASNetMobile, despite its compact size, had a slower inference time (108.67 ms), and slightly lower accuracy (90.94%). Performance differences among the models were statistically significant ( p < 0.0001). Conclusions : DenseNet121 demonstrated a superior diagnostic performance, while MobileNetV2 provided the most efficient solution for deployment in resource-constrained settings. The CNNs show substantial potential for improving melanoma detection in clinical and mobile applications.

Context: Guideline-directed medical therapy of heart failure (HF) primarily targets neurohormonal activation. However, growth hormone (GH) has emerged as a potential treatment for the multiple hormonal deficiency syndrome, which is associated with worse outcomes in HF., Objective: This study evaluates the efficacy and safety of GH therapy in HF., Data Sources: A systematic search was conducted in PubMed, Cochrane Library, and ClinicalTrials.gov, according to PRISMA guidelines., Study Selection: Randomized, placebo-controlled trials studying GH therapy in adult HF patients were included. Of the 1,184 initially identified records, 17 studies (1.4%) met the inclusion criteria., Data Extraction: Two independent authors conducted the search, with any disagreements resolved by a third author. Study quality was assessed using predefined criteria, including randomization, blinding, and the presence of a placebo group., Data Synthesis: A random-effects model was applied due to heterogeneity across studies. GH therapy significantly improved left ventricular ejection fraction (3.34%, 95% CI: 1.09% to 5.59%, p=0.0037), peak oxygen consumption (2.84 mL/kg/min, 95% CI: 1.32 mL/kg/min to 4.36 mL/kg/min, p=0.0002), and New York Heart Association class (-0.44, 95% CI: -0.08 to -0.81, p=0.023). GH therapy also reduced the composite of death, worsening HF or ventricular tachycardia by 41% (95% CI: 0.39-0.90, p=0.013). Subgroup analyses indicated that patients with ischemic cardiomyopathy, baseline ejection fraction ≥30% and longer treatment duration experienced greater benefits., Conclusions: GH therapy may improve cardiac function, exercise capacity, and HF symptoms, with a trend towards improvement in hard endpoints, such as worsening HF. Event-driven trials are necessary to validate these findings., (© The Author(s) 2024. Published by Oxford University Press on behalf of the Endocrine Society.)

Artificial intelligence processes are increasingly being used in emergency medicine, notably for supporting clinical decisions and potentially improving healthcare services. This study investigated demographics, coagulation tests, and biochemical markers routinely used for patients seen in the Emergency Department (ED) concerning hospitalization. This retrospective observational study included 13,991 emergency department visits of patients who had undergone biomarker testing to a tertiary public hospital in Greece during 2020. After applying five well-known classifiers of the caret package for machine learning of the R programming language in the whole data set and to each ED unit separately, the best performance regarding AUC ROC was observed in the Pulmonology ED unit. Furthermore, among the five classification techniques evaluated, a random forest classifier outperformed other models.

Emergency department (ED) overcrowding is an increasing global problem raising safety concerns for the patients. Elaborating an effective triage system that properly separates patients requiring hospital admission remains difficult. The objective of this study was to compare a clustering-related technique assignment of emergency department patients with the admission output using the k-means algorithm. Incorporating such a model into triage practice could theoretically shorten waiting times and reduce ED overcrowding.

Multidrug resistant infections in intensive care units represent a worldwide problem with adverse health effects and negative economic implications. As artificial intelligence techniques are increasingly applied in diagnosing, treating, and preventing antimicrobial resistance, in this study, we explore the use of association rule mining in the antibiotic resistance profile of critically ill patients suffering from multidrug resistant infections.

Association rule mining is a very popular unsupervised machine learning technique for discovering patterns in large datasets. Patients with stone disease commonly suffer from urinary tract infections (UTI), complicated by the emergence of antimicrobial resistance (AMR), due to the excessive use of antibiotics. In this study, we explore the use of association rule mining in the AMR profile of patients suffering from stone disease.

Benign prostatic enlargement (BPE) is a common disease in men over 50 years old. The phenotype of patients with BPE is heterogenous, regarding both baseline patient characteristics and disease-related parameters. Treatment can be either medical-conservative or surgical. A great variety of surgical techniques are available for surgical management, with three of the most common being monopolar transurethral resection of the prostate (mTUR-P), bipolar transurethral resection of the prostate (bTUR-P), and bipolar transurethral vaporization of the prostate (bTUVis). The selection of each one of these depends on surgeon reasoning, equipment availability, patient characteristics, and preferences. Since all of these techniques are available in our Urology Department, and surgeons are skilled to perform each one of them, we performed a clustering analysis according to patient pre-operative characteristics, using the k-means algorithm, to compare clustering-related technique assignment with the real-life technique used.

The objective of this study was to evaluate the predictive capability of five machine learning models regarding the admission or discharge of emergency department patients. A Random Forest classifier outperformed other models with respect to the area under the receiver operating characteristic curve (AUC ROC).

Emergency ambulance use is deemed necessary for the transport of acutely ill patients to hospital emergency departments (ED). However, some patients are discharged as they present low acuity or chronic problems and should receive primary healthcare services, while the most severely ill are admitted. In the present study, we examined the descriptive epidemiology of ambulance transports for emergencies in the ED by utilizing the data of the information systems of a public tertiary general hospital in Greece. More than half of the patients transferred to the ED by an ambulance were finally admitted to the hospital (52.25%), whereas only one-third (33.74%) of those transferred by other means. A statistically significant association was detected between ambulance use and hospital admission. Age was also statistically significantly higher in the ambulance group. Higher mean values of creatinine, CRP, LDH, urea, white-blood-cell count, and neutrophils were detected in the ambulance group, in contrast to hemoglobin and lymphocyte count which were higher in the non-ambulance group.

The objective of this study was to establish a machine learning model and to evaluate its predictive capability of admission to the hospital. This observational retrospective study included 3204 emergency department visits to a public tertiary care hospital in Greece from 14 March to 4 May 2019. We investigated biochemical markers and coagulation tests that are routinely checked in patients visiting the Emergency Department (ED) in relation to the ED outcome (admission or discharge). Among the most popular classification techniques of the scikit-learn library through a 10-fold cross-validation approach, a GaussianNB model outperformed other models with respect to the area under the receiver operating characteristic curve.

Data sharing among different entities in the healthcare domain has become an increasingly common practice, where each entity would most likely want to prevent indirect data disclosure via inference channels. The Local Distortion Hiding (LDH) algorithm has been developed to protect sensitive decision tree (DT) rules, which are chosen not to be disclosed when DT construction techniques are applied to the data. This article presents eight experiments using a Java-based prototype that implements the LDH algorithm in a diabetes data set. Our experiments test the ability of the LDH algorithm in two ways, firstly in inference control and secondly in maintaining the structure and the performance metrics of the resulting DT. Our experiments on hiding eight terminal nodes in a diabetes data set using a Java-based prototype that implements the LDH algorithm, yield satisfactory results.

During the COVID-19 pandemic, the number of visits in emergency departments (ED) worldwide decreased significantly based on several studies. This study aims to compare the patient flow in the emergency surgery department during the COVID-19 pandemic and a control period in the emergency department of a public tertiary care hospital in Greece. The overall patient flow reduction regarding the ED visits between the two examined periods was 49.07%. The emergency surgery department's corresponding visits were 235 and 552, respectively, which indicated an overall patient flow decrease of 57.43%. Chi-square analysis showed that age groups and ambulance use had statistically significant associations with the periods examined. An independent samples t-test was applied and deduced that the average patient's age was statistically significantly higher in the COVID-19 pandemic than in the non-pandemic period. By analyzing hospital information system data, useful conclusions can be drawn to prepare a surgical emergency unit better and optimize resource allocation in a healthcare facility in similar critical situations.