Improved pediatric ICU mortality prediction for respiratory diseases: machine learning and data subdivision insights.

The growing concern of pediatric mortality demands heightened preparedness in clinical settings, especially within intensive care units (ICUs). As respiratory-related admissions account for a substantial portion of pediatric illnesses, there is a pressing need to predict ICU mortality in these cases. This study based on data from 1188 patients, addresses this imperative using machine learning techniques and investigating different class balancing methods for pediatric ICU mortality prediction. This study employs the publicly accessible "Paediatric Intensive Care database" to train, validate, and test a machine learning model for predicting pediatric patient mortality. Features were ranked using three machine learning feature selection techniques, namely Random Forest, Extra Trees, and XGBoost, resulting in the selection of 16 critical features from a total of 105 features. Ten machine learning models and ensemble techniques are used to make accurate mortality predictions. To tackle the inherent class imbalance in the dataset, we applied a unique data partitioning technique to enhance the model's alignment with the data distribution. The CatBoost machine learning model achieved an area under the curve (AUC) of 72.22%, while the stacking ensemble model yielded an AUC of 60.59% for mortality prediction. The proposed subdivision technique, on the other hand, provides a significant improvement in performance metrics, with an AUC of 85.2% and an accuracy of 89.32%. These findings emphasize the potential of machine learning in enhancing pediatric mortality prediction and inform strategies for improved ICU readiness.

Systemic lupus erythematosus is associated with increased risk of mortality and acute kidney injury in patients with COVID-19 hospitalization: Insights from a National Inpatient Sample analysis.

INTRODUCTION: The COVID-19 pandemic has significantly impacted global health, especially for patients with chronic diseases that may compromise the immune system. This study investigates the association between systemic lupus erythematosus (SLE) and COVID-19 outcomes. METHODS: Data from the National Inpatient Sample (NIS) were analyzed to create a retrospective cohort of COVID-19 hospitalizations, comparing patients with and without SLE. Propensity-score matched analysis was conducted to assess the association between SLE and clinical outcomes in COVID-19 hospitalizations. RESULTS: The study included over a million COVID-19 hospitalizations, with approximately 0.5% having a secondary diagnosis of SLE. The SLE-COVID hospitalizations were predominantly female and younger, with a median age of 57.2, while the non-SLE-COVID group had a median age of 64.8 years. Comorbidities such as chronic obstructive pulmonary disease, renal failure, liver disease, and others were more prevalent in the SLE-COVID group. Patients with SLE and COVID-19 had a significantly higher incidence of acute kidney injury requiring dialysis than those without SLE. In-hospital mortality was higher in the SLE group, particularly in the 18-44 year age group (6.15% vs 2.47%, p = .022). CONCLUSION: COVID-19 patients with SLE are at an increased mortality risk, especially in the younger age group, and a higher incidence of acute kidney injury requiring dialysis. The elevated risk of adverse outcomes underscores the vulnerability of SLE patients to COVID-19. These findings emphasize the importance of special precautions and patient education for individuals with SLE to mitigate the risks associated with COVID-19.

Deep Learning Framework for Liver Segmentation from T1-Weighted MRI Images.

The human liver exhibits variable characteristics and anatomical information, which is often ambiguous in radiological images. Machine learning can be of great assistance in automatically segmenting the liver in radiological images, which can be further processed for computer-aided diagnosis. Magnetic resonance imaging (MRI) is preferred by clinicians for liver pathology diagnosis over volumetric abdominal computerized tomography (CT) scans, due to their superior representation of soft tissues. The convenience of Hounsfield unit (HoU) based preprocessing in CT scans is not available in MRI, making automatic segmentation challenging for MR images. This study investigates multiple state-of-the-art segmentation networks for liver segmentation from volumetric MRI images. Here, T1-weighted (in-phase) scans are investigated using expert-labeled liver masks from a public dataset of 20 patients (647 MR slices) from the Combined Healthy Abdominal Organ Segmentation grant challenge (CHAOS). The reason for using T1-weighted images is that it demonstrates brighter fat content, thus providing enhanced images for the segmentation task. Twenty-four different state-of-the-art segmentation networks with varying depths of dense, residual, and inception encoder and decoder backbones were investigated for the task. A novel cascaded network is proposed to segment axial liver slices. The proposed framework outperforms existing approaches reported in the literature for the liver segmentation task (on the same test set) with a dice similarity coefficient (DSC) score and intersect over union (IoU) of 95.15% and 92.10%, respectively.

On the qualitative study of a two-trophic plant-herbivore model.

The coexistence of plant-herbivore populations in an ecological system is a fundamental topic of research in mathematical ecology. Plant-herbivore interactions are often described by using discrete-time models in the case of non-overlapping generations: such generations have some specific time interval of life and their old generations are replaced by new generations after some regular interval of time. Keeping in mind the dynamical reliability of continuous-time models we presented two discrete-time plant-herbivore models. Mainly, by applying Euler's forward method a discrete-time plant-herbivore model is obtained from a continuous-time plant-herbivore model. In addition, a dynamically consistent discrete-time plant-herbivore model is obtained by applying a nonstandard difference scheme. Moreover, local stability is discussed and the existence of bifurcation about positive equilibrium is shown under some mathematical conditions. To control bifurcation and chaos, a modified hybrid technique is developed. Finally, to support our theocratical results and to show the dynamical reliability of the nonstandard difference scheme some numerical examples are provided.

A Dynamically Consistent Nonstandard Difference Scheme for a Discrete-Time Immunogenic Tumors Model.

This manuscript deals with the qualitative study of certain properties of an immunogenic tumors model. Mainly, we obtain a dynamically consistent discrete-time immunogenic tumors model using a nonstandard difference scheme. The existence of fixed points and their stability are discussed. It is shown that a continuous system experiences Hopf bifurcation at one and only one positive fixed point, whereas its discrete-time counterpart experiences Neimark-Sacker bifurcation at one and only one positive fixed point. It is shown that there is no chance of period-doubling bifurcation in our discrete-time system. Additionally, numerical simulations are carried out in support of our theoretical discussion.

Machine learning classification of texture features of MRI breast tumor and peri-tumor of combined pre- and early treatment predicts pathologic complete response.

PURPOSE: This study used machine learning classification of texture features from MRI of breast tumor and peri-tumor at multiple treatment time points in conjunction with molecular subtypes to predict eventual pathological complete response (PCR) to neoadjuvant chemotherapy. MATERIALS AND METHOD: This study employed a subset of patients (N = 166) with PCR data from the I-SPY-1 TRIAL (2002-2006). This cohort consisted of patients with stage 2 or 3 breast cancer that underwent anthracycline-cyclophosphamide and taxane treatment. Magnetic resonance imaging (MRI) was acquired pre-neoadjuvant chemotherapy, early, and mid-treatment. Texture features were extracted from post-contrast-enhanced MRI, pre- and post-contrast subtraction images, and with morphological dilation to include peri-tumoral tissue. Molecular subtypes and Ki67 were also included in the prediction model. Performance of classification models used the receiver operating characteristics curve analysis including area under the curve (AUC). Statistical analysis was done using unpaired two-tailed t-tests. RESULTS: Molecular subtypes alone yielded moderate prediction performance of PCR (AUC = 0.82, p = 0.07). Pre-, early, and mid-treatment data alone yielded moderate performance (AUC = 0.88, 0.72, and 0.78, p = 0.03, 0.13, 0.44, respectively). The combined pre- and early treatment data markedly improved performance (AUC = 0.96, p = 0.0003). Addition of molecular subtypes improved performance slightly for individual time points but substantially for the combined pre- and early treatment (AUC = 0.98, p = 0.0003). The optimal morphological dilation was 3-5 pixels. Subtraction of post- and pre-contrast MRI further improved performance (AUC = 0.98, p = 0.00003). Finally, among the machine-learning algorithms evaluated, the RUSBoosted Tree machine-learning method yielded the highest performance. CONCLUSION: AI-classification of texture features from MRI of breast tumor at multiple treatment time points accurately predicts eventual PCR. Longitudinal changes in texture features and peri-tumoral features further improve PCR prediction performance. Accurate assessment of treatment efficacy early on could minimize unnecessary toxic chemotherapy and enable mid-treatment modification for patients to achieve better clinical outcomes.

Brain MR Image Enhancement for Tumor Segmentation Using 3D U-Net.

MRI images are visually inspected by domain experts for the analysis and quantification of the tumorous tissues. Due to the large volumetric data, manual reporting on the images is subjective, cumbersome, and error prone. To address these problems, automatic image analysis tools are employed for tumor segmentation and other subsequent statistical analysis. However, prior to the tumor analysis and quantification, an important challenge lies in the pre-processing. In the present study, permutations of different pre-processing methods are comprehensively investigated. In particular, the study focused on Gibbs ringing artifact removal, bias field correction, intensity normalization, and adaptive histogram equalization (AHE). The pre-processed MRI data is then passed onto 3D U-Net for automatic segmentation of brain tumors. The segmentation results demonstrated the best performance with the combination of two techniques, i.e., Gibbs ringing artifact removal and bias-field correction. The proposed technique achieved mean dice score metrics of 0.91, 0.86, and 0.70 for the whole tumor, tumor core, and enhancing tumor, respectively. The testing mean dice scores achieved by the system are 0.90, 0.83, and 0.71 for the whole tumor, core tumor, and enhancing tumor, respectively. The novelty of this work concerns a robust pre-processing sequence for improving the segmentation accuracy of MR images. The proposed method overcame the testing dice scores of the state-of-the-art methods. The results are benchmarked with the existing techniques used in the Brain Tumor Segmentation Challenge (BraTS) 2018 challenge.

Primary intracranial extraosseous Ewing's sarcoma.

INTRODUCTION: Common sites of occurrence of extraosseous Ewing's sarcoma include the soft tissues and bones of the lower extremity, 12 paravertebral, and retroperitoneal regions. Primary intracranial Ewing's sarcoma/pPNET is usually intraparenchymal located 13 when supratentorially, and an extraaxial epidural tumor radiographically mimicking a meningioma is extremely rare. CASE PRESENTATION: A 20-year14 old male presented to the emergency department with a 1-day history of drowsiness, headache, and fever. Neurological exam15 ination revealed decreased muscle strength (4/5) in the left lower limb. Head computed tomography scan showed an epidural 16 space-occupying lesion in the right temporoparietal region, which was assumed to be a meningioma by radiographic criteria. However, the surgical specimen was diagnosed as Ewing's sarcoma. CONCLUSION: Primary intracranial extraosseous Ewing's sarcoma is a rare condition that may mimic a meningioma on imaging. Physicians must be cognizant of this possibility, particularly in any young individual with a solitary contrast-enhancing dural-based lesion.

DEW: A wavelet approach of rare sound event detection.

This paper presents a novel sound event detection (SED) system for rare events occurring in an open environment. Wavelet multiresolution analysis (MRA) is used to decompose the input audio clip of 30 seconds into five levels. Wavelet denoising is then applied on the third and fifth levels of MRA to filter out the background. Significant transitions, which may represent the onset of a rare event, are then estimated in these two levels by combining the peak-finding algorithm with the K-medoids clustering algorithm. The small portions of one-second duration, called 'chunks' are cropped from the input audio signal corresponding to the estimated locations of the significant transitions. Features from these chunks are extracted by the wavelet scattering network (WSN) and are given as input to a support vector machine (SVM) classifier, which classifies them. The proposed SED framework produces an error rate comparable to the SED systems based on convolutional neural network (CNN) architecture. Also, the proposed algorithm is computationally efficient and lightweight as compared to deep learning models, as it has no learnable parameter. It requires only a single epoch of training, which is 5, 10, 200, and 600 times lesser than the models based on CNNs and deep neural networks (DNNs), CNN with long short-term memory (LSTM) network, convolutional recurrent neural network (CRNN), and CNN respectively. The proposed model neither requires concatenation with previous frames for anomaly detection nor any additional training data creation needed for other comparative deep learning models. It needs to check almost 360 times fewer chunks for the presence of rare events than the other baseline systems used for comparison in this paper. All these characteristics make the proposed system suitable for real-time applications on resource-limited devices.

Enhancing Cutting Efficiency and Minimizing Forces for Nomex Honeycomb Core Using Grey Relational Analysis and Desirability Function Analysis.

Nomex Honeycomb core is the foundational building block for manufacturing aerospace composite components. Its usage requires machining honeycomb in complex aerodynamic profiles where the quality of the core is governed by accuracy and precision of cut profiles. The assessment of accuracy and precision is directly related to forces induced in the cutting tool and cutting efficiency. These two parameters form the basis of a multi-objective function that this paper aims to optimize for the milling operation. The parameter of depth of cut considered in this paper has not been analyzed in a multi-objective optimization study of the Nomex Honeycomb core previously. A Taguchi-based array of Design of Experiments followed by Analysis of Variance and correlation analysis is utilised. The results indicate that the most significant factor is the feed rate, with a percentage contribution of 72% for the cutting forces and depth of cut, with a percentage contribution of 85% in the case of cutting efficiency. The two parameters are optimized using Desirability Function Analysis and Grey Relational Analysis. The results are validated through experimental runs with an error within 5% of the statistical predictions, with the percentage improvement in cutting forces for optimum runs as compared to the worst experimental run at 47.8%. The percentage improvement in cutting efficiency likewise is 11%.

Effect of implantable loop recorder-based continuous rhythm monitoring on incident atrial fibrillation and stroke: An updated systematic review and meta-analysis of randomized controlled trials.

BACKGROUND: In patients with a history of stroke or those at high risk for developing stroke, a continuous rhythm monitoring strategy using an implantable loop recorder (ILR) is often performed to screen for atrial fibrillation (AF). OBJECTIVES: The purpose of this study was to perform a systematic review (MEDLINE and EMBASE) including randomized controlled trials comparing ILR-based continuous rhythm monitoring vs usual care in patients with a history of stroke or patients at high risk for developing stroke. METHODS: A meta-analysis was performed, and aggregate risk ratio (RR) and risk difference (RD) with 95% confidence interval (CI) were calculated. RESULTS: Four randomized controlled trials with 7237 patients (ILR 2114, non-ILR 5123) were included. ILR vs non-ILR was associated with increased detection of incident AF (RR 3.88; 95% CI 2.23-6.75; P <.00001; number needed to treat [NNT] = 7.7; I2 = 61%), increased appropriate initiation of anticoagulation (RR 2.29; 95% CI 2.07-2.55; P <.00001; NNT = 6.7; I2 = 0), and a 25% lower risk of developing stroke (RR 0.75; 95% CI 0.59-0.95]; P = .02; NNT = 100; I2 = 0%). In patients with history of stroke there was no difference in the risk of developing incident stroke (RR 0.83; 95% CI 0.61-1.14]; P = .25; I2 = 0%). CONCLUSION: Our meta-analysis showed that screening for AF with ILR is associated with increased detection of AF and increased initiation of appropriate anticoagulation therapy in patients with a history of stroke or those with risk factors for stroke. The benefit of stroke risk reduction with ILR remains unclear, and future studies focused on the inclusion of patients without a history of stroke are needed to elucidate this uncertainty.

Outcomes of COVID-19 Disease in Comparison with Influenza in Renal Transplant Recipients: Results from a Large Nationwide Research Network in the United States.

COVID-19 infection has worse outcomes in immunocompromised individuals. This includes those with diabetes mellitus, cancer, chronic autoimmune diseases requiring immunomodulatory therapy, and solid-organ transplant recipients on chronic immunosuppression. Using the National Inpatient Sample Database, this study retrospectively compared 14,915 renal transplant recipients who were hospitalized with either COVID-19 or Influenza virus infection in the US at any point between 1st January 2020 and 31st December 2020. We found that compared to renal transplant recipients with influenza infection, recipients with COVID-19 infection were more likely to require mechanical ventilation and vasopressor support and develop acute kidney injury requiring hemodialysis. COVID-19 patients also had significantly longer length of hospital stay. Renal transplant recipients with COVID-19 had significantly higher in-hospital mortality compared to recipients with influenza infection (14.09% vs 2.61%, adjusted odds ratio [aOR] 9.73 [95% CI (5.74-16.52)], P < .001). Our study clearly demonstrates the severe outcomes of high mortality and morbidity in renal transplant recipients with COVID-19. Further research should be undertaken to focus on the key areas noted to reduce morbidity and mortality in this population.

Anti-Inflammatory, Antipyretic, and Analgesic Potential of Chitin and Chitosan Derived from Cockroaches (Periplaneta americana) and Termites.

The chitin and chitosan biopolymers are extremely valuable because of their numerous industrial and pharmacological uses. Chitin and chitosan were extracted from the exoskeleton of Periplaneta americana (cockroaches) and termites using various acid and alkali techniques. The extraction process involves an initial demineralization step, during which integument dry powder was subjected to 500 mL (2.07 mol/L) of concentrated HCl at 100 degrees Celsius for 30 min, followed by meticulous rinsing with distilled water to restore the pH to its baseline. Deproteinization was conducted at 80 degrees Celsius using 500 mL (1 mol/L) of NaOH solution, which was repeated for 24 h. A total of 250 mL (0.06 mol/L) of NaOH was added at 100 degrees Celsius for 4 h to obtain chitosan, followed by extensive washing and subsequent drying. FTIR analysis was used to identify the functional groups in Periplaneta americana and termites. The crystallinity of these biopolymers, which have a face-centered cubic structure, was determined by X-ray diffraction analysis. This study assessed the analgesic properties of chitin and chitosan via an acetic-acid-induced writhing test in mice, revealing a significant reduction in writhing behavior following the chitin and chitosan extract. Notably, chitin exhibits the highest degree of analgesic activity compared to chitosan. Both chitin and chitosan show anti-inflammatory effects, with chitosan absorbing proton ions at sites of inflammation, while chitin effectively inhibits ear edema and elicits an analgesic response in mice. Furthermore, the present study revealed antipyretic activity, with termite chitin demonstrating the most significant effect at a concentration of 500 µL/mL, followed by chitosan and chitin at 100 µL/mL. These findings indicate the potential of using chitin and chitosan derived from termites and Periplaneta americana as natural anti-inflammatory compounds, implying prospective uses in anti-inflammatory, antipyretic, and analgesic capabilities.

Sex differences in atrial fibrillation ablation outcomes in patients with heart failure.

BACKGROUND: There is a lack of data on the impact of sex on the outcomes of patients with heart failure (HF) undergoing atrial fibrillation (AF) ablation. We aimed to analyze the association of sex with outcomes of atrial fibrillation ablation in patients with heart failure. METHODS: The National Readmissions Database (NRD) was analyzed from 2016 to 2019 to identify patients ≥ 18 years old with heart failure (HF) undergoing AF ablation. The outcomes of interest included peri-procedural complications, in-hospital mortality, resource utilization, and unplanned 1-year readmissions. The final cohort was divided into patients with HFrEF and HFpEF and outcomes were compared between males and females in both cohorts. RESULTS: A total of 23,277 patients with HF underwent AF ablation between 2016 and 2019, of which 14,480 had HFrEF and 8,797 had HFpEF. Among patients with HFrEF, 61.6% were males and 38.4% were females whereas, among patients with HFpEF, 35.4% were males and 64.6% were females. On a multivariable-adjusted analysis, in patients with HFrEF, there was no difference in the odds of in-hospital mortality, peri-procedural complications, or 1-year HF-related/AF-related/all-cause readmissions between males and females. In patients with HFpEF, females had a higher risk 1-year HF-related readmissions (adjusted hazards ratio: 1.46; 95% CI: 1.13-1.87; p = 0.01), without any difference in the 1-year AF-related/all-cause readmissions, in-hospital mortality, or peri-procedural complications. CONCLUSION: Our results show that females with HFrEF undergoing AF ablation have similar outcomes whereas females with HFpEF have higher 1-year HF readmissions with no difference in the other outcomes, compared to males.

Colon and Rectal Surgery Fellowship Applicant and Matching Trends in the United States: An Analysis of National Resident Matching Program Data.

BACKGROUND: Colon and Rectal Surgery fellowships are training programs that aim to train surgeons in the management of small bowel, colon, rectal, and anal pathologies. OBJECTIVE: We investigated trends in Colon and Rectal Surgery fellowship match to help applicants anticipate future fellowship application cycles. DESIGN: This was a retrospective cohort study of applicants in the Colon and Rectal Surgery match from 2009 to 2023. Proportion of positions filled, match rates, and rank-order lists were collected. The impact of US-MD, non-US-MD, and DO status on match rate was assessed. We used the Mann Kendall trend test to obtain tau statistic and P-value for temporal trends over time, while associations between categorical variables were investigated by a chi-square test. RESULTS: Fellowship programs increased from 43 to 67, positions increased from 78 to 110, and number of applicants rose from 113 to 135. Nearly all positions were filled from 2009 to 2023 (range: 96.3%-100%). The overall match rate fluctuated between 67.3% and 80.7%. The match rate over the past 5 years was 72.0%. The match rate for US-MDs was 80.0%, while non-US-MDs had a 56.2% match rate. The percentage matching at each rank were first choice 28.0%, second choice 10.4%, third choice 6.9%, and fourth choice or lower 23.5%. CONCLUSION: Despite an increase in Colon and Rectal Surgery fellowship positions, the overall match rate has not changed significantly over the years, mainly as a result of increased applicants.

Sonographic Characteristics and Pathology Correlation of Breast Imaging Reporting and Data System (BI-RADS) Category 4 Lesions.

INTRODUCTION: The Breast Imaging-Reporting and Database System (BI-RADS) category 4 is designated for breast lumps that do not display the typical features of malignancy but still raise enough suspicion to warrant a recommendation for a biopsy, as malignancy cannot be ruled out through imaging alone. The main objective of this study was to investigate the sonographic characteristics and pathology correlation of BI-RADS 4 breast lesions and determine the positive predictive rate of BI-RADS 4 lesions in diagnosing breast cancer, using histopathology as the gold standard. METHODS: This was a cross-sectional study conducted at the Department of Radiology, Aga Khan University Hospital in Karachi, spanning from May 2021 to August 2022, with a duration of 15 months. The study focused on female patients over the age of 18 who presented with suspicious breast lesions on ultrasound. Both mammography and ultrasound-guided core needle biopsy were performed on these patients, followed by a detailed histopathological evaluation of the biopsy specimens. To calculate the positive predictive value (PPV), true positive cases were identified through both histopathology and ultrasonography. RESULTS: A total of 227 cases were categorized as BI-RADS 4 lesions, with the patients' mean age being 47.8 ± 14.3 years (range: 17 - 88). Among the biopsied lesions, 101 cases were confirmed to be true positive for breast malignancies, resulting in a PPV for malignancy of 44.9%. Conversely, there were 124 false positive cases out of the 227 BI-RADS 4 category lesions (54.63%). The primary indication for presentation was a breast lump, and out of the 101 confirmed malignant cases, 70 (69.3%) were associated with malignancy. CONCLUSION: BI-RADS 4 can be utilized to assess suspicious breast lumps; however, for more reliable results and to avoid false negatives, histopathological confirmation should complement the imaging findings.

Self-Attention MHDNet: A Novel Deep Learning Model for the Detection of R-Peaks in the Electrocardiogram Signals Corrupted with Magnetohydrodynamic Effect.

Magnetic resonance imaging (MRI) is commonly used in medical diagnosis and minimally invasive image-guided operations. During an MRI scan, the patient's electrocardiogram (ECG) may be required for either gating or patient monitoring. However, the challenging environment of an MRI scanner, with its several types of magnetic fields, creates significant distortions of the collected ECG data due to the Magnetohydrodynamic (MHD) effect. These changes can be seen as irregular heartbeats. These distortions and abnormalities hamper the detection of QRS complexes, and a more in-depth diagnosis based on the ECG. This study aims to reliably detect R-peaks in the ECG waveforms in 3 Tesla (T) and 7T magnetic fields. A novel model, Self-Attention MHDNet, is proposed to detect R peaks from the MHD corrupted ECG signal through 1D-segmentation. The proposed model achieves a recall and precision of 99.83% and 99.68%, respectively, for the ECG data acquired in a 3T setting, while 99.87% and 99.78%, respectively, in a 7T setting. This model can thus be used in accurately gating the trigger pulse for the cardiovascular functional MRI.

Heart Failure Emergency Readmission Prediction Using Stacking Machine Learning Model.

Heart failure is a devastating disease that has high mortality rates and a negative impact on quality of life. Heart failure patients often experience emergency readmission after an initial episode, often due to inadequate management. A timely diagnosis and treatment of underlying issues can significantly reduce the risk of emergency readmissions. The purpose of this project was to predict emergency readmissions of discharged heart failure patients using classical machine learning (ML) models based on Electronic Health Record (EHR) data. The dataset used for this study consisted of 166 clinical biomarkers from 2008 patient records. Three feature selection techniques were studied along with 13 classical ML models using five-fold cross-validation. A stacking ML model was trained using the predictions of the three best-performing models for final classification. The stacking ML model provided an accuracy, precision, recall, specificity, F1-score, and area under the curve (AUC) of 89.41%, 90.10%, 89.41%, 87.83%, 89.28%, and 0.881, respectively. This indicates the effectiveness of the proposed model in predicting emergency readmissions. The healthcare providers can intervene pro-actively to reduce emergency hospital readmission risk and improve patient outcomes and decrease healthcare costs using the proposed model.