Medical data analysis

Context. Accurate mortality prediction in Intensive Care Units (ICUs) is essential for patient management and resource allocation. While structured clinical data (vitals, labs, severity scores) are widely used, unstructured text such as discharge summaries contains additional prognostic information often overlooked. Aims. This study evaluates whether discharge summaries alone, without structured variables, can predict 30-day post-discharge mortality with high accuracy. I tested recurrent deep learning models under different word embedding strategies. Methods. I created a balanced dataset of 30,000 summaries (15k deaths within 30 days, 15k survivals) from MIMIC-III (MIMIC-III 2016). Preprocessing included lowercasing, punctuation/number removal, stopword filtering, and synonym-based augmentation (Wei and Zou 2019). Tokenization was limited to 20,000 words, with sequences truncated/padded to 300 tokens. I trained four models: LSTM and BiLSTM with custom embeddings, and LSTM and GRU with GloVe embeddings (Wang et al. 2021). All models used Adam optimizer, dropout, and early stopping. Results. The BiLSTM with custom embeddings achieved the highest accuracy (87%). Compared to other models, it reduced false positives while maintaining balanced recall. The LSTM with custom embeddings reached higher recall but generated more false alarms. These findings highlight precision-recall trade-offs important for clinical decision-making. Conclusions. Discharge summaries can serve as standalone predictors of post-ICU mortality. Domain-specific embeddings and bidirectional architectures improve performance. Future work should integrate structured data, test transformer-based models (e.g. Clini-calBERT (Alsentzer et al. 2019)), and explore explainability tools to increase clinical trust.

Целью работы является разработка метода прогноза тяжести инфекции и выбора вида дыхательной поддержки пациентов с COVID-19. Решаются задачи классификации исходного состояния и течения болезни у пациентов с инфекцией COVID-19 и построения математической модели динамики состояний пациентов в отделении интенсивной терапии. В работе анализируются данные анамнеза, оценивается влияние сопутствующих хронических заболеваний и возраста пациента на тяжесть COVID-19 и эффективность терапии. Построена и исследована математическая модель динамики состояний пациентов с COVID-19. Получены оценки параметров модели для групп пациентов с разными хроническими заболеваниями. Уточнена структура индекса коморбидности 1 , позволяющего оценить влияние сопутствующих хронических (коморбидных) заболеваний на тяжесть и исход COVID-19. Предложен подход к выбору эффективного метода поддержки дыхания у пациентов с тяжелыми формами инфекции COVID-19. Ключевые слова: COVID-19, математическое моделирование, марковская модель, коморбидность.

Addressing the effects of class imbalance on feature selection models has become an increasingly important focus in academic research. This study introduces a novel support vector machine (SVM)-based algorithm specifically designed to handle class imbalance during the feature selection process. Using the Taiwan bankruptcy dataset as a case study, the algorithm incorporates the ExtraTreeClassifier() to manage class imbalance and identify a reduced set of relevant variables. To validate the selected features, SVM is applied within the imbalanced data context. Subsequently, analysis of variance (ANOVA) ranking is employed to further refine the variable set to three key features. An SVM model tailored for class imbalance is then constructed to assess the effectiveness of the final feature set. The proposed model significantly outperforms existing approaches in terms of classification performance. Specifically, it achieves a Type I error of 1.17% and a Type II error of 22.9%, compared to 4.4% and 39.4% reported in prior research. In terms of overall accuracy, our method reaches 83.1%, surpassing the 81.3% achieved by earlier studies. These results demonstrate that the proposed feature selection algorithm not only improves SVM accuracy but also outperforms other feature selection techniques when used in conjunction with SVMs, particularly under conditions of class imbalance.

Predictive machine learning (ML) and Artificial Intelligence (AI) have become revolutionary in healthcare and have the potential to identify diseases early, create individualised interventions, optimise operations, and increase health equity. The outlined paper analyses the history of predictive ML in healthcare, its main applications along with ethical implications of using it. Based on peer-reviewed sources and practical applications, it summarizes the perspective of predictive ML as the factor to improve the quality of life of individuals and populations and define the challenges that have to be overcome to lead to sustainable adoption.

Personalized Medicine is of paramount interest for many areas in Medical Informatics. Therefore genotype data as well a phenotype data about patients have to be available. This data will be stored in Electronic Health Records or -patient controlled -in Personal Health Records. As the amount of (raw) data is rising continuously, methods for a secure data administration have to be found. Grid Services offer data storage, can support data retrieval and the presentation of the data. The basic security services could be provided by the German health professional infrastructure, but there are many security challenges to be faced.

As artificial intelligence (AI) becomes increasingly integrated into wellness program systems, it promises transformative benefits, from early disease detection to personalized remedy. whatever, alongside these advantages arise ethical concerns and shifts in patient perspectives. This paper explores the intersection of AI, health resources ethics, and patient experience, critically analyzing the implications of algorithmic decision-making on medical practice, trust, equity, and autonomy. By examining current technologies, ethical frameworks, and patient responses, the paper argues for a human-centered approach to AI integration in health protection

Nowadays Parkinson's disease has been discovered that approximately 94% of people suffer from voice disorder problems. A neurodegenerative can identify PD patients through examination and multiple scanning tests. So, it usually takes more time to diagnose the disease at the early stage. Current work has identified that speech disorders can be a significant signal for Parkinson's disease. Therefore, this work proposed a fusion model to identify the speech disorder at the starting stage of the disease. In this process, the author has tested a model with a different pattern of feature selection method as well as classification mode and created a system with the best pattern. For the creation of pattern, three types of feature selection methods namely Chi-square, genetic algorithm and Embedded random forest method and four classifier models such as KNN, Naïve Bayes, SVM, Decision tree and Random Forest have been utilized. To analyze the performance of the system speech public dataset from the UCI repository,the authors applied the combination of the Embedded random feature selection method and random forest classification algorithm provides 97.89% of accuracy. However, this outcome is better than the recent work. The SMOTE is utilized for the balancing of the dataset.

During the last years we collected data of abdominal septic shock patients from clinics all over Germany. The mortality of septic shock is about 50%. Septic shock is related to immune system reactions and unusual measurements. Septic shock patients are intensely medicated during their stay at the intensive care unit. To help physicians recognizing the critical states of their patients as early as possible, we built a rule based alarm system based on a neuro-fuzzy inference machine. Analysing the patient data in a time window, we show the time dependency of the classification results. We give detailed classification results and explanation by rules. The results are compared to results obtained by using the most common scores in intensive care medicine. We discuss the advantages of the paradigms "neural networks" and "scores", and we answer the important question: Is a neural network more performant than scores for abdominal septic shock patient data?

The healthcare industry is experiencing a paradigm shift driven by the exponential growth of data generated from electronic health records (EHRs), medical imaging, genomic sequencing, and wearable technologies. Traditional data processing methods are insufficient to harness the full potential of these vast datasets. This research investigates the transformative role of Big Data Analytics (BDA) and Data Mining (DM) techniques powered by Artificial Intelligence (AI) in improving healthcare outcomes. The study employs a mixed-methods approach, combining a review of recent AI applications with a performance evaluation of machine learning models applied to anonymized healthcare datasets. Techniques such as classification, clustering, and predictive modeling are analyzed using tools like Python, Apache Spark, and TensorFlow. Key findings demonstrate that AI-enhanced BDA can significantly improve disease prediction accuracy, enable early diagnosis, optimize treatment plans, and support clinical decision-making. Moreover, precision medicine initiatives benefit from pattern recognition in large patient cohorts, enabling personalized therapies. However, challenges such as data privacy, algorithmic bias, and system interoperability persist. This research highlights the critical need for ethical, scalable, and interpretable AI solutions in healthcare and suggests pathways for future research and policy development.

Breast cancer is a significant health concern within medical care systems, necessitating accurate classification. The patient data are recorded and statistically analyzed, revealing an increasing number of files. And then transferred to the statistics department with increasing numbers. This study investigates breast cancer data imbalance utilizing Khartoum State Hospital. An imbalanced data problem occurs when one class has a significantly larger number of samples than another. To address this, resampling, attribute selection, handling missing values, classifier algorithms (ANN, REP TREE, SVM, J48), and ensemble learning models were employed. The base classifier yielded the first result, the meta-learning algorithms (Bagging, Boosting, and Random Subspace) the second, and an ensemble model the third. The boosting with the J48 ensemble model achieved the highest accuracy, 95.2797 %, outperforming bagging with j48 (90.559%) and random subspace with j48 (84.2657%).

This review talks about how Artificial Intelligence will bring about a sea change in medical diagnosis, particularly in imaging, genomics, and personalized medicine. AI algorithms are developed to increase the speed and accuracy of interpretation of medical images like X-rays and MRIs, greatly improving diagnostic accuracy and allowing rapid interventions. AI also supports radiologists by outlining areas of concern, reducing human error, and optimizing workflow efficiency. The applications of AI in genomics bring out the exact identification of genetic mutations, which can help a doctor derive a treatment plan based on the needs of each patient. However, there are many challenges to integrating AI into healthcare, such as possible biases, integration problems, and high costs that may increase healthcare disparities. Unless these challenges are addressed, there could be unequal creation and access to AI-driven diagnostic tools, and their potential for improving patient outcomes will not be completely realized.

This review talks about how Artificial Intelligence will bring about a sea change in medical diagnosis, particularly in imaging, genomics, and personalized medicine. AI algorithms are developed to increase the speed and accuracy of interpretation of medical images like X-rays and MRIs, greatly improving diagnostic accuracy and allowing rapid interventions. AI also supports radiologists by outlining areas of concern, reducing human error, and optimizing workflow efficiency. The applications of AI in genomics bring out the exact identification of genetic mutations, which can help a doctor derive a treatment plan based on the needs of each patient. However, there are many challenges to integrating AI into healthcare, such as possible biases, integration problems, and high costs that may increase healthcare disparities. Unless these challenges are addressed, there could be unequal creation and access to AI-driven diagnostic tools, and their potential for improving patient outcomes will not be completely realized.

This paper presents several aspects of the migration of a former implementation of a prestige Project ("ICPC2000") toward the new Microsoft.NET technology. This project and its extensions has as a result an integrated primary care information system that allow information storing/retrieving and exchange for a GP network and the related entities (County Health Insurance Houses -CHIH, hospitals, laboratories, pharmacies etc.). The former results are described, in order to understand the structure and the functions of the integrated information system suggested. Then, the new structure is described, with emphasizis on the specific aspects this approach: WEB services. Several considerations about the migration are presented and conclusions are issued.

Artificial Intelligence (AI) is transforming the drug development and Clinical Trials by improving efficiency, accuracy, and decision-making. AI predicts Pharmacokinetic (PK) and Pharmacodynamic (PD) properties, automates compound screening and enhances clinical testing throughput. In trial design, AI optimizes patient stratification and outcome prediction by analyzing vast datasets from previous trials and electronic health records, leading to cost-effective and adaptive trials. AI also facilitates real-time data monitoring, identifying discrepancies early to ensure data integrity and regulatory compliance. By integrating diverse data sources it streamlines clinical operations, reducing human error and manual workload. However, challenges persist in data quality and integration due to varying standards across sources, necessitating advanced harmonization techniques. Regulatory frameworks often lag behind AI advancements, creating uncertainty and potential delays. Ethical concerns, including patient privacy and data security, must also be addressed for responsible AI implementation. Establishing standardized protocols and ensuring regulatory alignment are critical for AI's successful integration into clinical research. In conclusion, AI revolutionizes drug development and clinical trials, enhancing efficiency and accuracy. However, overcoming data, regulatory, and ethical challenges is essential for its widespread adoption.

One of the most common causes of road accidents is driver behavior. To reduce abnormal driver behavior, it must be detected early on. Previous research has demonstrated that behavioral and physiological indicators affect drivers' performance. The goal of this study is to consider the feasibility of classifying driver behavior as either aggressive (sudden left or right turns, accelerating and braking), normal (average driving events) or slow (keeping a lower-than-average speed). Innovation in data mining and machine learning (ML) has allowed for the creation of powerful prediction tools. ML techniques have shown potential in predicting driver behavior, with classification being a critical study area. The data set was gathered using the Kaggle platform. This study classifies driver behavior using Orange3 data mining tools and tests several classifiers, including AdaBoost, CN2 rule inducer, and random forest (RF) classifiers. The results showed that AdaBoost was superior in predicting driver behavior, with 100% accuracy, while the classification accuracy in CN2 rule inducer and RF was 99.8% and 95.4%, respectively. These results demonstrate the possibility of early and highly accurate driver behavior prediction and use it to create a ML-based driver behavior detection system.

This paper presents a unifying framework for applying data analytics in the finance and healthcare sectors, where large-scale datasets demand robust and domain-specific methods. Drawing on machine learning and statistical techniques, the study addresses challenges such as data heterogeneity, stringent privacy regulations, and real-time analytics requirements. By comparing supervised and unsupervised approaches, the framework demonstrates how to integrate model outputs into operational and strategic decision-making pipelines. Emphasis is placed on managing compliance, ensuring data security, and achieving interpretability of analytic results. The paper ultimately provides guidance for industry professionals and researchers seeking to harness advanced analytics for tasks like fraud detection, risk assessment, patient outcomes analysis, and resource allocation, thereby advancing data-driven decision-making in both fields.

Background: Cardiovascular diseases (CVDs) continue to pose a critical public health challenge, contributing to approximately 31% of global deaths, according to the World Health Organization (WHO). Among these, hypertensive and ischemic heart diseases are particularly prevalent and often go undetected until serious complications arise. Traditional diagnostic procedures, though accurate, are often time-consuming, resourceintensive, and inaccessible to many, especially in remote and underserved regions. Moreover, the reliance on clinical judgment and subjective interpretation can sometimes lead to delayed or missed diagnoses. With the increasing availability of healthcare data and computational tools, machine learning has emerged as a powerful approach for developing intelligent systems that can assist in the early detection and prevention of chronic illnesses. By analysing large volumes of patient data and identifying hidden patterns, ML-based models have the potential to enhance clinical decision-making, reduce diagnostic errors, and improve patient outcomes. However, despite promising research, the real-world implementation of such systems remains limited, and many existing models lack proper deployment or validation across diverse populations. Addressing this gap, our study introduces CardioCare, a machine learning-powered heart disease prediction tool that not only offers high predictive accuracy but also ensures practical usability through a web-based interface, bridging the divide between academic models and accessible digital healthcare solutions. Materials and Methods: In this prospective the study utilized the Cleveland Heart Disease dataset from the UCI repository, containing 14 clinical features from 303 patient records. After cleaning and preprocessing the data through encoding and normalization, the dataset was split into training and testing sets (80:20). Various supervised machine learning models, including Logistic Regression, K-Nearest Neighbours, SVM, and Random Forest, were applied. Performance was evaluated using accuracy, precision, recall, and F1-score. The Random Forest classifier outperformed others, achieving an accuracy of 85%. To support practical usability, the final model was deployed through a Flask-based web application for real-time heart disease risk prediction. Results: Among the evaluated machine learning models, the Random Forest Classifier achieved the highest performance with an accuracy of 85%, precision of 83%, recall of 84%, and F1-score of 83.5%. It outperformed Logistic Regression, K-Nearest Neighbour, and Support Vector Machine in terms of both consistency and predictive capability. The model demonstrated strong generalization on unseen data. These results validate the model's effectiveness in classifying heart disease presence based on key clinical attributes.

In this paper, we compare the efficiency of two binary classifiers. The first one uses the Weighted Ordered Weighted Averaging (WOWA) aggregation function whose coefficients are learned thanks to a genetic algorithm. The second is based on an artificial neural network trained by a backpropagation algorithm. They are trained to be used in a multi-criteria decision system. These kind of multi-criteria system are more and more common in the cyber-defence field. In this work, we compare the performance of these two classifiers by using two criteria: Area Under the Curve of a Receiver Operating Characteristics (ROC) curve and the Area Under the Curve of a Precision-Recall (P-R) curve. This second criterion is more adapted for imbalanced dataset what is often the case in the cyber-security field. We perform a complete parameter study of these classifiers to optimize their performance. The dataset used for this work is a pool of Hypertext Preprocessor (PHP) files analyzed by a multiagent PHP webshell detector. We obtain different good results, especially for neural networks and highlights the advantage of the genetic algorithm method that allows a physical interpretation of the result.

In this paper, we compare the efficiency of two binary classifiers. The first one uses the Weighted Ordered Weighted Averaging (WOWA) aggregation function whose coefficients are learned thanks to a genetic algorithm. The second is based on an artificial neural network trained by a backpropagation algorithm. They are trained to be used in a multi-criteria decision system. These kind of multi-criteria system are more and more common in the cyber-defence field. In this work, we compare the performance of these two classifiers by using two criteria: Area Under the Curve of a Receiver Operating Characteristics (ROC) curve and the Area Under the Curve of a PrecisionRecall (P-R) curve. This second criterion is more adapted for imbalanced dataset what is often the case in the cyber-security field. We perform a complete parameter study of these classifiers to optimize their performance. The dataset used for this work is a pool of Hypertext Preprocessor (PHP) files analyzed by a multiagent PH...

AI-driven solutions are transforming healthcare quality-of-care ratings by addressing challenges such as fragmented data, inconsistent scoring, and reliance on manual processes. Traditional rating systems incorporate diverse measures, including clinical outcomes, adherence to best practices, and patient experiences, but face inefficiencies due to interoperability issues and data silos. AI offers a unified approach by integrating and standardizing data from multiple sources, enabling automated analysis, real-time monitoring, and predictive insights. AI-driven sentiment analysis further enhances objectivity by processing unstructured patient feedback. Implementing AI in quality assessments can improve accuracy, provider accountability, and patient outcomes while reducing costs. To maximize its potential, healthcare organizations must invest in AI infrastructure, enforce standardized protocols, train staff, and ensure ethical data use. Collaboration among healthcare entities will further refine AI-driven assessments, advancing a patient-centered, data-driven approach to quality ratings.

Due to its decentralised, accessible, and secured structure, blockchain-a framework that has historically been viewed with suspicion-has developed into a groundbreaking breakthrough. Reliable automated scripting and static data records are also a part of this system. The intricacy and expertise required to operate with blockchain technology is one of its disadvantages. Machine Learning (ML) frameworks are the key fixing this problem. It has the potential to make Blockchain more user-friendly. In this research paper, we reviewed the current state of blockchain-based cloud computing, machine learning, and safe data exchange. In this study, we suggest using a blockchain technology to provide a safe space where health entities can share medical information transparently. A modified advanced encryption standard (MAES) based blockchain encryption model is developed by us. This method is based on engineering and AI principles, and its main objective is to design useful artefacts that improve their respective domains.

The integration of Artificial Intelligence in healthcare represents a transformative advancement in modern medicine, particularly in diagnostic applications and personalized treatment approaches. This comprehensive article examines the current state and future prospects of AI in healthcare, focusing on medical diagnosis, data analytics, and precision medicine. The article encompasses various applications, from imaging diagnostics and pattern recognition to personalized treatment planning, highlighting significant improvements in diagnostic accuracy, workflow efficiency, and patient outcomes.

AI technologies have rapidly advanced, bringing transformative changes to various industries and aspects of daily life. However, the rise of AI has also raised significant ethical concerns regarding fairness, transparency, accountability, and privacy. This chapter delves into the development of responsible and transparent machine learning models, emphasizing the importance of ethical considerations in real-world applications.

Education institutions and teachers are in desperate need of automated, nonintrusive means of getting student feedback that would allow them to better understand the learning cycle and assess the success of course design. Students would benefit from a framework that intelligently guides their actions and provides exercises or resources to support and enhance their learning. The recommender system framework is a software agent that learns the user's preferences through a variety of channels and then utilizes that knowledge to provide product suggestions. A recommendation engine considers all potential user interests as background information, uses that knowledge to produce convincing recommendations, and then returns those ideas to the user. This article presents a feature selection and machine learning based course recommendation system for higher education students. principal component analysis (PCA) algorithm is used for feature selection. AdaBoost, k nearest neighbour (KNN), and Naïve Bayes algorithms are used to classify and predict student data. It is found that the AdaBoost algorithm is having better accuracy and F1 score for course recommendation to students. PCA AdaBoost is achieving an accuracy of 99.5%.

The rapid advancement of Artificial Intelligence (AI) in healthcare has revolutionized disease prediction, early diagnosis, and preventive medicine. Traditional methods of disease detection often rely on subjective assessments and late-stage symptoms, leading to delayed treatment and increased mortality rates. AI, through machine learning (ML) algorithms, deep learning networks, and big data analytics, is enhancing diagnostic accuracy by identifying patterns in vast medical datasets. This paper explores how AI is transforming disease prediction across various fields, including oncology, cardiology, endocrinology, and neurology. AI-driven models such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Support Vector Machines (SVMs) are significantly improving early detection rates of cancer, heart disease, diabetes, and mental health disorders. Moreover, AI-powered predictive analytics from electronic health records (EHRs), wearable devices, and genomic sequencing are enabling healthcare professionals to make data-driven decisions for personalized treatment plans. Despite its potential, AI in disease prediction faces challenges such as algorithmic bias, data privacy concerns, and ethical implications regarding decision-making autonomy. This paper examines recent advancements in AI-based disease prediction, its clinical applications, ethical considerations, and future prospects. The findings highlight how AI is shaping the future of precision medicine, reducing healthcare costs, and improving patient outcomes through early disease detection and preventive care strategies.

Big data mainly refers to a huge volume of rapidly
growing data over size exabytes (1018). A major chunk of this
data is unstructured text data produced from several sources.
In order to use such data effectively, they need to be processed
and organized. Taxonomy, a hierarchical structure, is
considered an effective way of organizing the data. In the past,
many techniques have been proposed to generate taxonomy
automatically. Recently some attempts have also been made to
evolve the static structure of taxonomy to deal with the rapidly
changing nature of data. However, the voluminous nature of
today’s data currently exceeds the processing capabilities of
conventional techniques. In this regard, there is a need for a
scalable technique that potentially speeds up the process of
taxonomy generation and evolution and caters to a large
amount of unstructured big data. This paper presents a
technique for both the generation and the evolution of
taxonomy on the Apache Spark framework. The technique is
tested on a text dataset belonging to a computing domain. The
test results show that the scalable taxonomy generation and
evolution technique proposed in this paper is not only timeefficient
but also produces a good quality taxonomy as
compared to state-of-the-art techniques.

Ever since the rise of human civilization, more and more diseases have been discovered with the rapid growth of medical knowledge. This sheer volume of information makes it hard for humans to memorize or even utilize it efficiently. Thus, machine learning emerged as a powerful tool for complex calculations by offering a solution to this challenge. This paper intends to use deep learning and machine learning algorithms to develop a predictive model that can recognize potential diseases based on symptoms. The model is then seamlessly integrated into a text-based disease prediction assistant chatbot that serves as a communication platform between the users and the system. The algorithms researched for the disease prediction models are knearest neighbours (KNN), support vector machines (SVM), random forest, and neural networks. After that, a chatbot application is created by integrating long short-term memory (LSTM), natural language toolkit (NLTK) libraries, and Telegram. As a result, the SVM models demonstrated excellent performance by achieving an accuracy of 92.24%, closely followed by random forest with 92.23%, KNN with 91.57%, and artificial neural network (ANN) with 91.52% accuracy. In short, this paper presents a potential solution for a more accurate disease prediction tool by implementing the best disease prediction model with the chatbot models together.

Broad informed consent is a special form of informed consent designed to allow hospitals to use medical data and patient residual biological samples for a wide range of applications in medical research. It is suitable for clinical research with low risk and uncertain research purpose. Applications of broad informed consent in hospitals include biobank management, in vitro diagnostic kit clinical trials, hospital teaching and research, and data use of hospital big data platforms. Despite the challenges of implementing it, broad informed consent contributes to scientific progress and public health. In this paper, we define the elements of broad informed consent, outline the challenges of its application in hospitals, and explore broad informed consent implementation. We believe that broad informed consent plays an important role in the development of medical research. This paper aims to improve the implementation efficiency of broad informed consent through implementation plans and measures to motivate medical personnel.

The integration of machine learning (ML) into health information technology (HIT) is revolutionizing data-driven healthcare systems, yet several key challenges and areas of focus remain. Electronic health records (EHRs) constitute most of the data source (60%), with wearable devices and interviews/focus groups comprising smaller portions. This indicates a continued reliance on traditional health records, although emerging technologies are beginning to play a role. Another key preprocessing challenge, with data cleaning consuming the most effort (40%), followed by data anonymization and feature selection, each requiring substantial effort in ensuring the accuracy and privacy of patient data. Supervised learning dominates in healthcare applications, followed by deep learning and unsupervised learning. In terms of accuracy, EHR data consistently yields the highest performance, around 85%, closely followed by wearable devices, genetic data, and lifestyle data. However, challenges remain in addressing data privacy and algorithm transparency, as highlighted by the distribution of effort in ensuring compliance and maintaining data privacy. The findings suggest a need for further exploration into wearable devices and the real-time monitoring capabilities they bring to healthcare, alongside tackling data preprocessing and ethical challenges in HIT.

The goal of this effort is to use the Gemma platform (Genetically Engineered Microbe for Medical Applications) framework is a family of lightweight, state-of-the art open models built from the same research and technology used to create the Gemini models to analyse illnesses and the symptoms associated with them. Gemma is an effective tool that finds patterns and links in massive datasets by applying sophisticated data analysis techniques. Applying Gemma to a vast set of medical records allowed us to find important new information on the symptomatology and symptoms of illness. GEMMA is a state-of-the-art technique that creates personalised microorganisms for medicinal applications by fusing genetic engineering with microbiology. Our results demonstrate Gemma's potential as a useful tool for doctors to diagnose and cure a range of ailments. By highlighting the value of applying data-driven methods for illness and symptom analysis and eventually enhancing patient outcomes, this research advances the area of healthcare.

Personalized medicine is transforming healthcare practices by tailoring specific treatment plans related to a patient's genetic profile, life patterns, and case history. The move to precision medicine finds its foundation in enhancements in the development of artificial intelligence and integration in health data. Generative AI, a type of AI trained from existing data to make new insights and predictions, has significant applications in analyzing complex data like genomic sequences, electronic health records, and realtime patient monitoring. AI integrated with health information will lead to earlier disease detection and better diagnostics, thus enabling customized treatment plans to improve the patient's outcomes and reduce health expenditures. This article will explore the currently ongoing technological transformation of healthcare, driven by AI and big data. It examines how personalized medicine empowers patients with real-time data and feedback, while various challenges keep arising with regard to data privacy, interoperability, and ethical concerns. It concludes with insight into the future of healthcare, where this segment of AI-driven personalized care has the possible propensity to transform how patients are treated.

The rapid advancement of Health Information Technology (HIT) has revolutionized cancer care, introducing new capabilities in early detection, individualized treatment, and data security. This paper reviews the key HIT innovations that have reshaped oncology, including the widespread adoption of Electronic Health Records (EHRs), the use of artificial intelligence (AI) in diagnostic processes, and the application of telemedicine and predictive analytics to improve treatment planning. These technologies have enabled earlier and more accurate diagnoses, expanded access to care, and facilitated data-driven, personalized treatment strategies. In parallel, this review addresses the importance of safeguarding sensitive patient data, focusing on encryption, anonymization techniques, and the regulatory frameworks like HIPAA and GDPR that ensure data privacy. Finally, the paper discusses future directions for HIT, particularly the potential of AI-driven decision support systems, real-time data analytics, and privacy-enhancing algorithms to further advance cancer care, creating a more secure, efficient, and patient-centered healthcare system.

The Internet of Medical Things (IoMT) has paved the way for innovative approaches to collecting and managing medical data. With the large and sensitive medical data being processed hence, the need for a strong identity and privacy become necessary. The present paper suggests a comprehensive method of PriMedGuard which aims at protection of the personal medical information. The first stage will be data collection from devices and sensors, then data cleaning to transform the data into the required format. There is also a safety system in the system that registers and authenticates authorized entities as well as ETDO (Enhanced Tasmanian Devil Optimization algorithm) is used for generating asymmetric cryptographic keys. The data is encrypted using the Secure Bit-Count Transmutation (SBCT) Data Encryption Algorithm and then put in the locations provided by the InterPlanetary File System (IPFS), a decentralized and distributed storage system. A safe smart contract on the blockchain is created so that the data retrieval is secure, and MedSecEnsemble Detection is proposed as an intrusion detection technique in the IoMT network. By using this method, data will stay available while at the same time integrity, confidentiality and protection against vulnerabilities are ensured. Hence, the Internet of Medical Things ecosystem will be secured from unauthorized access and possible security threats.

In Atlántico, Colombia, the Departmental Health Secretariat has been proactive in promoting healthy lifestyles to prevent noncommunicable diseases (NCDs), adhering to the strategies outlined in the Health Action Plan (PAS) as recommended by the Ministry of Health's Promotion and Prevention and Epidemiology and Demography Directorates. However, the efficacy of these activities is often hampered by a reliance on external data sources or studies from regions with dissimilar health determinants. This paper highlights the role of business intelligence tools-including dimensional fact models, multidimensional databases, ETL (Extract, Transform and Load) processes, and data visualization dashboards-in enhancing local data analysis, thereby providing valuable insights for public health management and research within Atlántico. By leveraging accurate, localized data, this approach strengthens decision-making processes in NCD prevention. The study aims to ascertain the impact of a healthy diet and physical activity on preventing NCDs among Atlántico's vulnerable populations, utilizing a confirmatory research methodology that employs observation, surveys, and interviews within the constraints of the COVID-19 pandemic. The findings underscore the critical role of diet and exercise in NCD prevention and demonstrate the efficacy of employing digital analytical tools and business intelligence to inform user-centric health interventions. This research lays the groundwork for developing evidence-based, locally tailored NCD prevention strategies, marking a significant advancement in public health initiatives in Atlántico.

Exploring patterns and trends in suicide data is crucial to understanding the complex interplay of factors that affect mental health and suicide risk. This study delves into the global suicide landscape using the "Suicide Data with Continent" dataset from Kaggle, which aggregates comprehensive information on suicides from various countries and regions. Using multivariate regression analysis along with sophisticated data analysis and visualization techniques, including trend charts, maps, and pie charts through the Dash and Plotly libraries, we uncover continental patterns and trends related to suicide rates. The analysis, based on a dataset comprising 27.820 records spanning 32 years with details on demographics, socioeconomic indicators, and 6.748.420 recorded suicides, aims to illuminate the multifaceted nature of suicide phenomena. The findings not only foster a deeper understanding of global suicide dynamics, but also serve as a foundational analysis to inform future suicide prevention policies and programs, specifically laying the groundwork for an impending study in Colombia.

In the evolving landscape of medical data analysis, clustering techniques play a pivotal role, particularly in deciphering intricate patterns within datasets, such as those linked to cancer diagnostics. With the continuous expansion and increasing complexity of healthcare data, there is a growing demand for effective clustering algorithms capable of extracting significant insights. Current trends underscore the necessity of carefully selecting the most appropriate clustering method to improve both the accuracy and interpretability of analytical results. In this paper, we conduct a comprehensive comparison of three prominent clustering algorithms-KMeans, Agglomerative Clustering, and Gaussian Mixture Model (GMM)-applied to a breast cancer dataset comprising features from Fine Needle Aspirates (FNA) of breast masses. Following a thorough preprocessing and scaling of the features, we assess the performance of these clustering techniques using the Silhouette Score, Calinski-Harabasz Score, and Davies-Bouldin Score. The findings reveal that KMeans provides superior cluster separation and clarity relative to the other algorithms. This research emphasizes the critical role of algorithm selection based on specific dataset attributes and evaluation metrics, aiming to enhance the accuracy of clustering outcomes in breast cancer classification.

Customer churn is a major challenge faced by e-commerce companies, as it leads to loss of revenue and decreased customer loyalty. In recent years, for predicting and reducing client churn machine learning techniques are powerful tools. This research aims to explore the use of machine learning algorithms for predicting customer churn, annual spending, and product ontime delivery in e-commerce. The study first conducted a comprehensive review of the literature on customer churn in machine learning. The literature showed that customer churn has been predicted successfully using a variety of machine learning algorithms, including support vector machine (SVM), random forest, and decision tree in various industries. To address this gap in the literature, the study conducted an empirical analysis of customer churn in e-commerce using machine learning algorithms. The data were then pre-processed and analyzed utilizing machine learning techniques for prediction. According to the study's findings, machine learning algorithms are effective in predicting customer churn, and product on-time delivery in e-commerce. The best-performing algorithm SVM achieved an accuracy of 83.45% in predicting customer churn and 68.42% for product on-time delivery prediction.

Collaborative efforts and research networks are essential for advancing AI in nephrology. Establishing partnerships between academic institutions, healthcare organizations, industry, and regulatory bodies can facilitate data sharing, standardization, and validation of AI models. Collaborative research initiatives, such as the Kidney Precision Medicine Project (KPMP), aim to integrate multi-omics data and AI to improve the understanding and treatment of renal diseases.

ABSTRACT⎯ Cardiovascular disease is the leading cause of death globally, claiming approximately 17.9 million lives each year according to the WHO. Cardiovascular diseases are a group of medical conditions that affect the heart and blood vessels, such as coronary heart disease, ischemic heart disease, valvular heart disease, heart failure, coronary artery disease, and peripheral vascular disease. More than 80% of deaths from cardiovascular diseases are caused by heart attacks and strokes, with one in three cases occurring prematurely in individuals under 70 years of age. This study aims to develop a predictive model to identify individuals at risk of heart attacks. In this study, several supervised machine learning algorithms were used, including logistic regression, neural networks, support vector machines, and decision trees. Stepwise logistic regression showed the best performance with an AUC of 0.91 on the testing data and 0.92 on the training data. The stepwise logistic regression model had high accuracy and good generalization ability with significant variables such as oldpeak, caa, sex, cp, restecg, slp, and thall. Neural networks and polynomial SVM experienced overfitting with poor performance on the testing data. Polynomial SVM with PCA was also less effective. For future research, it is recommended to further develop the stepwise logistic regression model and explore regularization and cross-validation techniques on neural networks and SVMs to reduce overfitting.


Keywords⎯ Cardiovascular, Decision Tree, Logistic Regression, Neural Network, Support Vector Machine.

ABSTRACT⎯ Cardiovascular disease is the leading cause of death globally, claiming approximately 17.9 million lives each year according to the WHO. Cardiovascular diseases are a group of medical conditions that affect the heart and blood vessels, such as coronary heart disease, ischemic heart disease, valvular heart disease, heart failure, coronary artery disease, and peripheral vascular disease. More than 80% of deaths from cardiovascular diseases are caused by heart attacks and strokes, with one in three cases occurring prematurely in individuals under 70 years of age. This study aims to develop a predictive model to identify individuals at risk of heart attacks. In this study, several supervised machine learning algorithms were used, including logistic regression, neural networks, support vector machines, and decision trees. Stepwise logistic regression showed the best performance with an AUC of 0.91 on the testing data and 0.92 on the training data. The stepwise logistic regression model had high accuracy and good generalization ability with significant variables such as oldpeak, caa, sex, cp, restecg, slp, and thall. Neural networks and polynomial SVM experienced overfitting with poor performance on the testing data. Polynomial SVM with PCA was also less effective. For future research, it is recommended to further develop the stepwise logistic regression model and explore regularization and cross-validation techniques on neural networks and SVMs to reduce overfitting.

Keywords⎯ Cardiovascular, Decision Tree, Logistic Regression, Neural Network, Support Vector Machine.

Ensemble learning, which involves combining the opinions of multiple experts to arrive at a better result, has been used for centuries. In this work, a review of the major voting methods in ensemble learning is explored. This work will focus on a new method for combining the results of individual learners. The method depends on the relative accuracy and diversity of teams. Instead of trying to assign weight to each different trainer, the concept of diversity teams is presented. Each team will vote as one player; however, the individual accuracies of each learner still be implemented. The concept of relaxing parameters that deal with each team as one player is presented. Our experiments demonstrate that traditional ensemble voting methods outperform individual learners. There is a limit to the superiority of the ensemble learner that any ensemble learner cannot go beyond. The proposed voting method gives the same results as the traditional ensemble voting methods, however, a different diversity of the proposed method from the traditional voting method or for different values of the relaxing parameter can be achieved.

The fields of artificial intelligence (AI) and machine learning (ML) have attracted significant interest and investment from a diverse range of industries, especially during the last several years. Despite the fact that AI methods have been used extensively and put through extensive testing in the healthcare industry, the recently discovered coronavirus disease (COVID-19) necessitates the use of these methods in order to prevent the emergence of the disease. The proposed system is based on six ML algorithms to predict COVID-19 infection as random forest (RF) algorithm, naive bayes (NB) algorithm, support vector machine (SVM) algorithm, decision tree (DT) algorithm, multi-layer perceptron (MLP), and k-nearest neighbor (KNN). It is based on two steps: first, we uploaded the dataset to train the model. Then, we test our model on those cases to work directly after making a trained classifier so it can directly discover with automatic COVID-19 prediction state of a patient suspected or not. The proposed system results showed the high accuracy of NB, DT, and SVM as 98.646%. Besides the better time to build the model and early predict the state of patients is 31 ms of the NB algorithm.

Artificial intelligence (AI) has emerged as a transformative force in healthcare, revolutionizing disease diagnosis and treatment planning. This article explores the growing role of AI in healthcare, focusing on its applications in various medical domains, such as radiology, oncology, and genomics. By leveraging advanced algorithms and machine learning models, AI systems can analyze vast amounts of medical data, identify patterns and anomalies, and assist healthcare professionals in making informed decisions. The article highlights the potential benefits of AI in facilitating early diagnosis, personalized medicine, and improved patient outcomes. It also discusses the advancements in AI technologies, including sophisticated deep learning models, processing and interpreting complex medical data, and integration with other technologies such as Internet of Things (IoT) and blockchain. However, the article also addresses the challenges associated with implementing AI in healthcare, such as ethical concerns, data privacy and security, the need for extensive and unbiased datasets, and the establishment of robust regulatory frameworks. The future outlook for AI in healthcare is promising, with ongoing advancements and growing interest from stakeholders. Nonetheless, the success of AI in healthcare relies on addressing the challenges, ensuring transparency and accountability, and maintaining a humancentered approach. By harnessing the power of AI responsibly and effectively, healthcare systems can pave the way for improved patient care and outcomes.

Diabetes mellitus is a hyperglycemia-like chronic condition that is a troublesome disease. It is estimated that, according to the growing morbidity, by 2040, the world will cross 642 million diabetic patients. This means that each one of the ten adults will be diabetes-affected. Diabetes can also lead to other illnesses such as heart attacks, kidney damage, and even blindness. The prediction of diabetes in advance motivates us to develop a machine learning-based model. A dataset was obtained from the online repository for this work. The obtained dataset was imbalanced. An imbalanced dataset presents a challenge that is needed to be balanced for prediction using multiple machine learning like Tomek and SMOTE. These techniques remove necessary outliers that are incomplete in the provided dataset. These outliers are also managed using the IQR method. Additionally, this research employed a two-stage model selection methodology. In the first stage, logistic regression, Support Vector Mac...

Data science  application  in  medical  education  is  rapidly  evolving.This  article  defined  data science  and  its  application  in  various  domains  of  mathematics,  statistics,  artificialintelligence, computer engineering and medical education. The concepts frequently deployed in  artificial  intelligence  such  as  deep  learning,  big  data  and  machine  learning  were explained.  A  brief  overview  of  evolution  of  data  science  was  narrated  and  breakthrough moments  were  equally  highlighted.  Medical  education  and  its  peculiarities  as  well  as  the various  challenges  encountered  in  Nigeria  were  examined.  The  value  of  integration  of medical  education  and  technology  along  with  the  attendant  benefits  was  highlighted. Appropriate recommendations were made on the need to leverage on data science to improve learning experience and outcomes in medical education.Keywords:Evolution, Data science, Artificial intelligence, Medical education, Application

"Bio Statistics can define as the application of mathematical tools used in statistics to field of biological science and medicine statistics. Collection, organization, analysis ,interpretation  and presentation of data very important part of bio medical research, which is mainly used in research studies .It deal with aspect of this , including the planning of data collection in term of the design of survey and experiments. Now a day, doing various statistical tests has been made easy by sophisticated computer software. But mainly two things are very important to researcher/ investigator is to choose the appropriate statistical test for the computer to perform based on the nature of data derived from one’s own research. The second is to understand if an analysis was performed appropriately during review and interpretation of others’ research. A basic understanding of biostatistics is needed to understand and interpret the medical literature. The basic step of research is defining the research question, review of literature, formulate hypothesis, preparing study design, data collection, analysis and interpretation of it.
Bio statistics is mainly concern with descriptive statistics and inferential statistics. The objective of this article is that to understand the type of data, choose the correct method of presentation and summarization of the data according to nature of data."