Healthcare analytics (New York, N.Y.)最新文献

The American healthcare system allocates considerable resources compared to peer-developed nations. However, outcomes significantly trail behind, particularly in life expectancy. This study addresses questions about the enduring trends in healthcare spending as a percentage of Gross Domestic Product (GDP), notable factors contributing to this concerning trend, and the timing to apply an emergency brake to curb this accelerating trajectory. Advanced machine learning algorithms, such as Random Forest and Support Vector Regression (SVR), in conjunction with traditional statistical forecasting methods, are used to forecast future patterns. The research underscores the importance of healthcare analytics in unraveling the intricacies of the healthcare system. The findings highlight the pressing need for effective policies to confront this mounting challenge.

Chronic liver disease (CLD) is a major health concern for millions of people all over the globe. Early prediction and identification are critical for taking appropriate action at the earliest stages of the disease. Implementing machine learning methods in predicting CLD can greatly improve medical outcomes, reduce the burden of the condition, and promote proactive and preventive healthcare practices for those at risk. However, traditional machine learning has some limitations which can be mitigated through ensemble learning. Boosting is the most advantageous ensemble learning approach. This study aims to improve the performance of the available boosting techniques for CLD prediction. Seven popular boosting algorithms of Gradient Boosting (GB), AdaBoost, LogitBoost, SGBoost, XGBoost, LightGBM, and CatBoost, and two publicly available popular CLD datasets (Liver disease patient dataset (LDPD) and Indian liver disease patient dataset (ILPD)) of dissimilar size and demography are considered in this study. The features of the datasets are ascertained by exploratory data analysis. Additionally, hyperparameter tuning, normalisation, and upsampling are used for predictive analytics. The proportional importance of every feature contributing to CLD for every algorithm is assessed. Each algorithm's performance on both datasets is assessed using k-fold cross-validation, twelve metrics, and runtime. Among the five boosting algorithms, GB emerged as the best overall performer for both datasets. It attained 98.80% and 98.29% accuracy rates for LDPD and ILPD, respectively. GB also outperformed other boosting algorithms regarding other performance metrics except runtime.

One of the realities of the COVID-19 worldwide pandemic is the occurrence of infected individuals with COVID-19 and two other diseases, Monkeypox and HIV. This study presents a compartmental deterministic epidemiological model with non-linear differential equations to study the transmission dynamics of the co-infection of the three diseases. Rigorous analysis of the model shows that the disease-free equilibrium was locally and globally asymptotically stable when the associated reproduction number of the diseases was not up to unity, showing that the spread of the diseases and their co-circulation can be effectively controlled in this circumstance. Real-life data about the diseases are collated and fitted to the model through which values of key parameters of the model were estimated. These parameters’ values were used to carry out numerical simulations of the model using MATLAB and validate the qualitative results obtained earlier from the model. The numerical simulation of the model was used to explore the interactions and dynamics resulting from the co-infection of COVID-19, HIV, and Monkeypox in humans, including the reciprocal impacts of each of the diseases on the other two, their patterns of coexistence and their effects on the host. We developed a tool to help predict the co-infection of the three diseases. Through the insights gained in this study, recommendations were made to policymakers in the healthcare sector on how to combat effectively and adequately the co-infection of the three diseases in the human population and mitigate their disease burden.

Neuronal cell segmentation identifies and separates individual neurons in an image, typically to study their properties or analyze their organization in the nervous system. This is significant because neurological problems and diseases can only be treated effectively when the structure and function of neurons are understood. The proposed method is based on convolutional neural networks (CNNs) and graph attention networks (GATs) for segmenting biomedical images. A contracting path built upon a couple of convolution layers and max pooling is included in the architecture to capture context. After that, the GATs are applied to the captured context. In GATs, each node in the graph is associated with a vector of hidden features, and the model calculates attention coefficients between pairs of nodes. These attention coefficients are learned during training and can be used to weigh the contribution of each node’s features to the representation of the graph. An expanding path that utilizes the outputs generated by GATs paves the way for exact segmentation. The dataset comprises 606 microscopic images, mainly categorized into different cell types (astrocytes, cortex, and SHSY5Y). By implementing our proposed U-GAT algorithm, we obtained the highest accuracy of 86.5% and an F1 score of 0.719 compared to the CNN, U-Net, SegResNet, SegNet VGG16, and GAT benchmarking algorithms. This proposed method could help researchers in the biotech industry develop novel drugs since a more accurate deep-learning method is essential for segmenting complex images like neuronal images.

This study presents the influence of COVID-19 and the pandemic on individuals diagnosed with hepatocellular carcinoma and intrahepatic cholangiocarcinoma, the two most common types of primary liver cancer. The study compares the effects before and after the pandemic on these patients. Additionally, it endeavors to predict the likelihood of survival for liver cancer patients. Our research will employ various methodologies to investigate this. Exploratory data analysis techniques are utilized, including univariate analysis, correlation analysis, bivariate analysis, chi-square testing, and T-sample testing. For predictive analytics, machine learning algorithms such as Logistic Regression, Decision Trees, Classification And Regression Tree (CART), Artificial Neural Networks (ANN), K-Nearest Neighbors (KNN), and Support Vector Machines (SVMs) will be applied. For our outputs, Logistic Regression and SVMs emerged as top-performing algorithms, boasting a remarkable accuracy rate of 93%. The study reveals that COVID-19 affected all age groups similarly. However, a gender-based difference was observed, indicating that males faced a higher risk of both cancer and mortality. Furthermore, the study found that variables such as year, month, bleeding, cirrhosis, and previously known cirrhosis did not significantly influence patient survival.

The post-COVID-19 landscape has propelled the global telemedicine sector to a projected valuation of USD 91.2 billion by 2022, with a remarkable compounded annual growth rate (CAGR) of 18.6% from 2023 to 2030. This paper introduces an analytical wearable healthcare monitoring device (WHMD) designed for the timely detection and seamless transmission of crucial health vitals to telemedical cloud agents. The fractional order modeling approach is employed to delineate the efficacy of the WHMD in pregnancy-related contexts. The Caputo fractional calculus framework is harnessed to show the device potential in capturing and communicating vital health data to medical experts precisely at the cloud layer. Our formulation establishes the fractional order model's positivity, existence, and uniqueness, substantiating its mathematical validity. The investigation comprises two major equilibrium points: the disease-free equilibrium and the equilibrium accounting for disease presence, both interconnected with the WHMD. The paper explores the impact of integrating the WHMD during pregnancy cycles. Analytical findings show that the basic reproduction number remains below unity, showing the WHMD efficacy in mitigating health complications. Furthermore, the fractional multi-stage differential transform method (FMSDTM) facilitates optimal control scenarios involving WHMD utilisation among pregnant patients. The proposed approach exhibits robustness and conclusively elucidates the dynamic potential of WHMD in supporting maternal health and disease control throughout pregnancy. This paper significantly contributes to the evolving landscape of analytical wearable healthcare research, highlighting the critical role of WHMDs in safeguarding maternal well-being and mitigating disease risks in edge reconfigurable health architectures.

Automated segmentation methods can produce faster segmentation of tumors in medical images, aiding medical professionals in diagnosis and treatment plans. A 3D U-Net method excels in this task but has high computational costs due to large model parameters, which limits their application under resource constraints. This study targets an optimized triplanar (2.5D) model ensemble to generate accurate segmentation with fewer parameters. The proposed triplanar model uses spatial and channel attention mechanisms and information from multiple orthogonal planar views to predict segmentation labels. In particular, we studied the optimum filter size to improve the accuracy without increasing the network complexity. The model generated output is further post-processed to fine-tune the segmentation results. The Dice similarity coefficients (Dice-score) of the Brain Tumor Segmentation (BraTS) 2020 training set for enhancing tumor (ET), whole tumor (WT), and tumor core (TC) are 0.736, 0.896, and 0.841, whereas, for the validation set, they are 0.713, 0.873, and 0.778, respectively. The proposed base model has only $10.25M$ parameters, three times less than BraTS 2020’s best-performing model (ET 0.798, WT 0.912, TC 0.857) on the validation set. The proposed ensemble model has $93.5M$ parameters, 1.6 times less than the top-ranked model and two times less than the third-ranked model (ET 0.793, WT 0.911, TC 0.853 on validation set) of BraTS2020 challenge.

Healthcare professionals must provide their patients with the best possible service and be well-informed and expert at carrying out complex surgical procedures to fulfill this responsibility. The aim of the medical treatments is fewer complications, shorter hospital stays, and a better patient experience. Through continuous learning and training, medical practitioners trained in up-to-date and state-of-the-art surgical techniques and technologies make productive and effective healthcare systems possible. Healthcare systems often report on problems with surgical processes, skipped procedures, unusual activities during operations, and lengthy transition times. This event log data allows implementing process mining methods to deliver medical professionals with simple and understandable findings using Petri nets for process analysis and enhancement. This study identifies the parallels and discrepancies between the pre-and post-stages and their respective frequency on each typical Central Venous Catheter (CVC) installation activity. The Process Mining for Python (PM4Py) frameworks used four major mining algorithms to view the event log (i.e., alpha miner, directly-follows graph (DFG), heuristic miner, and inductive miner). This study's findings indicate that medical residents are more susceptible to error during pre-operative procedures.

This study addresses a tumor–macrophage interaction model to examine the role of the saturated response of M2 macrophages. We find the equilibrium point of the model and analyze local stability at each equilibrium. We show that tumor-free equilibrium is always stable, whereas, under certain conditions, the tumor-dominant and interior equilibrium are asymptotically stable. Moreover, stable and unstable limit cycles and period-doubling bifurcation have been observed at the interior equilibrium point. A remarkable result has been observed: in the presence of a saturated response of M2 macrophages, with a relatively higher activation rate of M2 macrophages due to tumor cells, the disease spreads more quickly in the body. Hence, M1 macrophages cannot stabilize the system, and aperiodic oscillations are observed. Furthermore, we show that a better immune response can reverse that system’s unstable nature. Numerical simulations verify the analytical results.

Oral cancer is a form of cancer that develops in the tissue of an oral cavity. Detection at an early stage is necessary to prevent the mortality rate in cancer patients. Artificial intelligence (AI) techniques play a significant role in assisting with diagnosing oral cancer. The AI techniques provide better detection accuracy and help automate oral cancer detection. The study shows that AI has a wide range of algorithms and provides outcomes in the most precise manner possible. We provide an overview of different input types and apply an appropriate algorithm to detect oral cancer. We aim to provide an overview of various AI techniques that can be used to automate oral cancer detection and to analyze these techniques to improve the efficiency and accuracy of oral cancer screening. We provide a summary of various methods available for oral cancer detection. We cover different input image formats, their processing, and the need for segmentation and feature extraction. We further include a list of other conventional strategies. We focus on various AI techniques for detecting oral cancer, including deep learning, machine learning, fuzzy computing, data mining, and genetic algorithms, and evaluates their benefits and drawbacks. The larger part of the articles focused on deep learning (37%) methods, followed by machine learning (32%), genetic algorithms (12%), data mining techniques (10%), and fuzzy computing (9%) for oral cancer detection.