Security practices and insider threats in Spanish healthcare centers: a survey-based risk assessment.

Associations between maternal lipid metabolism and immune characteristics in children with autism spectrum disorder.

Objective To analyze spatial clusters and temporal trends regarding patients commuting to access cancer treatment in Brazil between 2015 and 2022. Methods This was a spatial analysis using data from hospital and outpatient information systems (2015-2022). Absolute and relative frequencies of treatments (surgery, chemotherapy, and radiotherapy) and commuting patterns were calculated. Cluster analysis (K-means) categorized distances into three intervals: low (2.1-261.4 km), medium (261.6-762.2 km), and high (764.0-3,865.8 km). Temporal trends were assessed by Prais-Winsten regression, estimating annual percentage change (β) and confidence intervals (95%CI) as a measure of dispersion. Results Of the 27,204,159 cancer services provided, 55.2% involved displacement to other municipalities. During the period, 3.6% of patients received surgical treatment, 7.1% received radiotherapy, and 89.3% received chemotherapy. There was a reduction in the distances traveled for hospitalization, from 93.0 km in 2015 to 84.2 km in 2022, with an annual decrease of 0.8% (95%CI -0.9; -0.7). For chemotherapy, the reduction was from 87.8 km to 83.5 km, with a variation of -0.4% per year (95%CI -0.4; -0.3). Distances for radiotherapy remained stable, with a slight variation of -0.3% (95%CI -0.9; 0.2). Conclusion The reduction in the distance traveled for hospitalization and chemotherapy contrasts with the stability in radiotherapy. The maintenance of long journeys for radiotherapy highlights disparities in the geographical distribution of these services. Thus, the urgency of decentralizing oncology services and investing in regional infrastructure is underscored to ensure access for the population, particularly those residing in rural and remote areas, and to guarantee equitable access to highly complex treatments.

Hospital-acquired infections among inpatients experiencing delayed hospital discharge: an observational study in Wales.

Objective To assess whether hospitalizations due to stroke were associated with Chagas disease and the Chagas disease vulnerability index in the state of Minas Gerais. Methods This was a population-based analytical observational study. Data were obtained from the Hospital Information System of the Brazilian National Health System in Minas Gerais for the year 2022. Hospitalization records were selected based on the International Statistical Classification of Diseases and Related Health Problems, in which stroke was recorded as the primary cause and Chagas disease/sentinels as the secondary cause. Stroke-related hospitalization rates were estimated, and a correlation analysis was conducted between the Chagas disease vulnerability index and stroke hospitalization prevalence in Minas Gerais and its health macro-regions. Results In 2022, a total of 25,813 hospitalizations due to stroke as the primary cause were recorded, of which only 18 included Chagas disease/sentinels as the secondary cause, most of them in the Central macro-region (n=10). The correlation between stroke-related hospitalizations and the Chagas disease vulnerability index was weak (r=0.145) and not statistically significant, despite the overlap of macro-regions with higher stroke-related hospitalization rates and higher Chagas disease vulnerability index scores (Jequitinhonha Valley and North). Conclusion Although there was a high number of stroke-related hospitalizations in the Brazilian National Health System in Minas Gerais, only 18 of these records were associated with Chagas disease/sentinels, which may suggest underreporting of Chagas disease and uncertainty regarding the quality of such records.

Background/aim: Positron emission tomography (PET)/CT using 18F-fluorodeoxyglucose (FDG) is a commonly used technique for staging and/or localization in various malignant diseases. Our study aimed to investigate the clinical laboratory and histopathological features of patients with incidental pituitary involvement. Materials and methods: In this retrospective study, approximately 60,000 18-FDG PET/CT scans performed for various indications between May 2012 and December 2020 were reviewed to identify patients with pituitary involvement. The clinical and laboratory characteristics of these patients were investigated using the Hospital Information Management System. Results: Twenty-five patients with incidental pituitary uptake were identified on 18F-FDG PET/CT scans. Eleven patients had macroadenomas. The mean SUVmax levels of the pituitary lesions and different malignancies were statistically similar (13.6 (5.40–38.80) and 11.7 (0.00–116.0), respectively. Six of the 25 patients had nonfunctional adenomas, one patient had a prolactinoma, and four patients had pituitary insufficiency based on hormonal evaluation. Two patients were found to have a pituitary adenoma on histopathological evaluation. One patient had a metastasis from lung carcinoma, as detected radiologically. Conclusion: Incidental pituitary uptake is often overlooked, particularly in the patients with advanced metastatic disease. In individuals with a primary malignancy who are subjected to significant physiological stress from chemotherapy, surgery, and other interventions, timely diagnosing and management of pituitary insufficiency is crucial, as it can significantly impact morbidity and mortality. To develop diagnostic algorithms, predictive models, or systematic follow-up strategies, prospective studies are needed that include hormonal, radiological, and histopathological follow-up data on both the primary malignancy and pituitary metastases.

Objectives: To examine the effectiveness of ice therapy in reducing acute pain in the early postoperative period in patients with sports-related knee injuries at the Pediatric Orthopedics Department of the Children’s Hospital Zagreb. Study Design: Cross-sectional study. Participants and Methods The research included a total of 84 examinees who underwent knee surgery. For the purposes of this research, data was collected from the Hospital Information System on examinees who underwent surgery after a knee sports-related injury between April 2022 and April 2023 at the Pediatric Orthopedics Department of the Children’s Hospital Zagreb, with regard to gender, age, diagnosis, type of surgery, the use of pain medication therapy, the use and duration of physical therapy with cold compresses, the length of postoperative treatment in the hospital, during which data on self-assessment of pain based on the visual-analogue scale (VAS) were collected every day. Results The most common cause of knee injuries was football (31%), with the most common diagnosis being patellar injury (24%). The most frequently performed surgical procedure was arthroscopy with reconstruction (64.3%). Out of all examinees who received ice therapy, 88.1% had a positive reaction, while 7.1% had a negative reaction. All of the examinees received therapy with cold compresses several times a day, and 33.3% of them received painkillers along with cold compresses. The most frequently used therapy was paracetamol (25%), followed by ketoprofen (19%) and ibuprofen (4.76%). Conclusion This study showed that painkillers are more effective in reducing the level of pain than the application of cold compresses. However, ice therapy significantly reduced postoperative pain both in girls and boys. Ice therapy and painkillers are most effective on the second postoperative day of treatment.

Among medical imaging systems that play a crucial role in modern medical diagnosis and treatment processes, X-ray imaging stands out as an essential diagnostic tool due to its low cost and wide accessibility. This study focuses on developing a model based on a Convolutional Neural Network (CNN) architecture to automatically identify and classify anatomical regions in X-ray images. Using the MURA dataset and the UNIFESP X-Ray Body Part Classification dataset obtained from Kaggle, detailed anatomical region and projection view classification was performed on 7,487 multi-view musculoskeletal radiographs. The classification process utilized the AlexNet and ResNet50 architectures. To enhance the transparency and interpretability of the decision mechanisms, visual analysis was conducted using the Grad-CAM technique on misclassified samples. The obtained results showed that the AlexNet model achieved a validation accuracy of 91.52%, while the ResNet50 model achieved 94.20%. These findings demonstrate that detailed anatomical and directional classification can be performed with high accuracy, suggesting that this method could serve as an effective approach to improving labelling accuracy in hospital information systems.

IntroductionCardiopulmonary bypass has been used to perform complex cardiac surgery for over 70 years. Advances in bypass techniques and perioperative medicine have increased the safety of cardiac procedures, leading to reduced morbidity and mortality. Nevertheless, cardiopulmonary bypass still carries risks, including systemic inflammation and dysfunction of various organs. To date, optimal blood pressure management during cardiopulmonary bypass remains a subject of ongoing debate. Conflicting evidence exists regarding negative outcomes associated with both low and high mean arterial pressures. Current clinical guidelines recommend a broad target range for mean arterial pressure during cardiopulmonary bypass, which underscores the existing gap in knowledge. In non-cardiac surgery, the time-weighted average of mean arterial pressure has been used to determine minimum safe thresholds, with greater deviation from 65 mm Hg associated with an increased risk of adverse outcomes. However, the definition and reporting of low blood pressure during cardiopulmonary bypass varies between studies, and the use of time-weighted averages below the threshold is still uncommon. Details on pump flow during extracorporeal circulation are seldom reported.Methods and analysisWe plan to conduct a retrospective, single-centre data analysis to investigate the effects of both arterial blood pressure and extracorporeal pump flow, including their time-weighted averages and areas under defined thresholds, during cardiopulmonary bypass on neurological outcomes in adult patients undergoing cardiac surgery between 2014 and 2023. The study will include both elective and emergency procedures, with separate analyses conducted based on the urgency and complexity of the operations. Digitally recorded anaesthesia and perfusion records will be imported and validated to extract information on haemodynamic parameters, neurological monitoring and extracorporeal circulation. Ischaemic and haemorrhagic strokes will be identified by screening postoperative brain imaging records for keywords indicating neurological events. Diagnostic data and additional patient and procedural information will be extracted from the local cardiac surgery database and hospital information system. Information about incidence and course of postoperative delirium will be extracted from the patient data management system used in intensive care. We expect to include approximately 500–700 cases per year in the final analysis.Ethics and disseminationThe local ethics committee approved our study (Ethics Committee of the Medical University of Graz, IRB00002556, 36-296 ex 23/24). We aim to publish the results of our study preferably in an open access format.Trial registration numberThe study protocol was registered at the Center for Open Science (https://doi.org/10.17605/OSF.IO/FAMV3).

In January 2019, the Chinese government launched a pilot National Volume-Based Procurement (NVBP) program, which was then gradually extended to the whole country to alleviate the financial burden of healthcare expenditure. This study aims to investigate the impact of the NVBP policy on healthcare expenditure. By extracting outpatient payment records of breast cancer patients in a tertiary hospital information system from May 2020 to April 2023, we performed an interrupted time series (ITS) to assess the impact of the NVBP policy on healthcare expenditure. ITS analyses showed that the NVBP policy mitigated the persistent increase in total average monthly per-visit expenditure, shifting it from an upward trend (9.68 CNY (p < 0.01)) to a flat trend (-2.33 CNY (p = 0.673)). The momentum of the continuous increase in the average monthly per-visit drug expenditure of patients has been effectively curbed; the expenditure has turned from an upward trend (15.10 CNY (p < 0.01)) to a downward trend (-11.19 CNY (p < 0.01)). Although the trend of continuous increase in average monthly per-visit expenditures for chemotherapy drugs (p < 0.01), oral endocrine drugs (p < 0.01) and Traditional Chinese Medicines (TCM) (p < 0.01) were effectively mitigated, especially for TCM; however, the expenditure of injectable endocrine drugs (p = 0.038), examination (p = 0.015) and laboratory (p = 0.029) were the persistent increases. Our study suggested that the NVBP policy effectively prevented the continued growth in drug expenditure for breast cancer outpatients, but did not substantially improve total expenditure. In addition, the policy has not only affected the expenditures of medicines, but has also changed the structure of expenditures.

The rapid digitization of healthcare has made Hospital Information Systems (HIS) essential for improving patient care, optimizing clinical workflows, and strengthening overall hospital management. This study examined HIS acceptance among 32 paramedical professionals including laboratory technicians, radiology staff, and pharmacists in a tertiary care hospital in Kerala, applying the model of Technology Acceptance Model (TAM) as the guiding framework. Primary data were collected through a structured questionnaire and informal interview and analyzed using descriptive statistics, percentage analysis, and multiple regression. Results showed strong perceived usefulness, with over 84% of respondents agreeing that HIS improves task efficiency and documentation accuracy, and 90% rating the interface as simple and user-friendly. Despite this, about one-quarter expressed neutral or low confidence in daily operations due to limited IT exposure, and 18.8 % reported difficulty obtaining timely technical support. Regression analysis revealed that attitude toward HIS was the strongest predictor of usage intention, explaining 46.4 % of the variance. These findings highlight the need for department specific training, regular refresher programs, and responsive technical assistance to build user confidence and ensure sustainable HIS adoption, offering practical guidance for hospital administrators and policymakers aiming to advance digital health initiatives.

ABSTRACT This paper explores how the perceived value of AI-integrated Hospital Information Systems (HIS) affects healthcare professionals’ self-efficacy and, hence, job satisfaction. Drawing on the Conservation of Resources (COR) theory, the study suggests that AI-enabled HIS, which includes wearable devices, chatbots, and machine learning, is a valuable informational resource that might lower stress, improve task mastery, and ease workload. This study examines how the perceived value of AI-based HIS (PVHIS) enhances healthcare workers’ self-efficacy in simplifying and automating their tasks, thereby improving job satisfaction. The study also tests digital leadership moderates the relationship between perceived HIS value and self-efficacy through strategic vision, technological mentoring, and fostering trust in AI systems. A quantitative survey of 403 healthcare professionals reveals that the perceived value of AI-based HIS positively predicts self-efficacy and job satisfaction and that digital leadership significantly strengthens the PVHIS–self-efficacy link. These results underscore digital leadership's critical role in determining effective AI integration in healthcare environments and promoting staff well-being, providing theoretical and pragmatic contributions.

PurposeThis study aimed to investigate the impact of shift characteristics, resident workload, and physician attributes on diagnostic expenditures in low-acuity patients (green triage category) emergency department (ED) patients.MethodsA retrospective cross-sectional analysis was conducted in a high-volume tertiary ED over one month, including 22,427 green-triage visits managed by 71 emergency medicine residents. Resident characteristics (age, gender, seniority, shift group, post-night shift status), patient demographics, and diagnostic expenditures (laboratory, imaging, electronic medication orders, total cost) were extracted from the hospital information system. Non-parametric tests were used for group comparisons given non-normal distribution patterns, and cost determinants were analyzed using a Gamma generalized linear model with a log-link function.ResultsDiagnostic expenditures demonstrated significant variability across physician and patient characteristics. Female residents were associated with higher laboratory (1.27 vs 0.87 USD; p<0.001), imaging (1.35 vs 1.01 USD; p<0.001), and medication-order costs (p<0.001). Compared with junior residents, mid-level trainees generated the highest total expenditures, whereas senior residents exhibited a cost-attenuating effect (exp(β)=0.74). Unadjusted analyses indicated greater total spending during night shifts (2.8 USD vs 2.39–2.43 USD; p=0.011); however, after adjustment for resident- and patient-level covariates, night-shift status was associated with lower expenditures (exp(β)=0.76). Post-night-shift status independently correlated with reduced laboratory and medication-order costs. Resident workload showed a strong inverse association with expenditures, with increasing daily patient volume linked to lower total diagnostic costs (rho=–0.226; p<0.001). Among patient factors, advancing age increased total cost by approximately 6% per year (exp(β)=1.06). Repeat ED utilization emerged as the most powerful cost determinant, with each additional prior visit associated with more than a threefold increase in diagnostic spending (exp(β)>3; p<0.001). Female patients consistently incurred higher costs across all categories (p<0.001).ConclusionDiagnostic spending in low-acuity ED encounters is shaped by both clinical and operational dynamics. Resident workload, seniority level, and gender independently influence cost patterns, while patient age and repeat admissions are strong drivers of increased expenditures. These findings highlight several potentially modifiable determinants—particularly workload distribution and trainee supervision—that may support more cost-conscious diagnostic practices in busy emergency departments.

Artificial intelligence-based follow-up systems offer efficient solutions for postdischarge care but face limited patient acceptance due to trust concerns. Trust is a key determinant of acceptance, yet existing instruments are insufficient for assessing trust in artificial intelligence-driven follow-up contexts. This study focused on developing and psychometrically validating a scale to assess patient trust in artificial intelligence-based follow-up. This methodological study was carried out in three sequential stages: item generation, scale development, and psychometric validation. During the initial phase, a preliminary item pool was established based on theoretical frameworks, existing literature, semistructured interviews, and focus group discussions. In the second phase, a preliminary scale was developed using the Delphi method, content validity assessment, and a pilot survey. In the third phase, a questionnaire was administered to patients who had experienced artificial intelligence-based follow-up. The structure was first determined by exploratory factor analysis (EFA), refined via item analysis, and then validated using confirmatory factor analysis (CFA). Reliability was assessed by computing internal consistency (Cronbach's α, McDonald's ω) and test-retest correlations. The final scale comprised 27 items across three dimensions: patient dispositional trust, system interaction trust, and environmental trust. Content validity was high (I-CVI: 0.833-1.000; S-CVI/Ave: 0.955). EFA revealed three factors that explained 66.0% of the variance. CFA indicated a good model fit (χ²/df = 1.096, RMSEA = 0.022, CFI = 0.999, TLI = 0.99). The reliability indices were strong (Cronbach's α = 0.931; McDonald's ω = 0.960; test-retest = 0.901). The Artificial Intelligence-Based Follow-Up Trust Scale demonstrated strong reliability and validity. This valuable tool can assess trust in artificial intelligence-based follow-up systems, supporting future research and the development of strategies to increase trust and improve follow-up adherence.

This study assessed digital health adoption and performance of healthcare services in Akure Metropolis, Ondo State, Nigeria, where there is an acute underutilization of the accessible technological tools. A survey research method was employed, enabling the collection of pertinent data from healthcare workers and patients attending the State Specialist Hospital, Akure, the largest healthcare facility in Akure metropolis, through questionnaire administration. The data was analysed using both descriptive and inferential statistics. The formulated hypotheses were tested with the use of test statistics, while Pearson product-moment correlation and factor analysis were used to test the level of relationship between the variables. The results identify the moderate application of SMS-based medication reminders and health education, as 63.5 per cent of the respondents stated that the cost of implementing and maintaining digital health technologies has a serious influence on the capacity of hospitals to provide quality care. Also, 63.0 per cent of the sampled population admitted that government policies and regulations are important in determining the reception of digital health. The research indicates low use of integrated Hospital Information Systems (HIS) and Laboratory Information Management Systems (LIMS) as well as automated billing, with 44.5 per cent of the respondents disagreeing with the statement that the training of healthcare professionals has a positive effect on service delivery. The regression model indicates that the two most significant independent contributors of digital health adoption are operation effectiveness and infrastructure readiness, and that operation effectiveness operated significantly and positively (beta = 0.757, p = 0.000). Conversely, policy, cost, and user readiness variables exerted a rather low effect, indicating that the development of digital health adoption rates should be directed to the improvement of operating systems, infrastructure, and the removal of organisational constraints. The study revealed that the healthcare system in Akure is technologically evolving, yet demonstrably capable of realising sizeable performance gains, where even limited digital tools are embedded.

The number of EMR users in RSUP Dr. M.Djamil Padang often occur disorders such as application errors, EMR hank or application can not be opened, employee/user complaints about the problem is not known to have been done. Purpose: This study aims to design a helpdesk application to streamline the reporting of electronic medical record issues, thereby addressing the existing problems. Methods: This study employed a Research and Development (R&amp;D) method. The research was conducted at Dr. M. Djamil Hospital Padang. Data were collected through observations and interviews with two key 2 informants: the head of the Hospital Information System (SIMRS) and one IT staff. A checklist table was also utilized. The application was designed using the waterfall development method. Results: he resulting helpdesk application for electronic medical record (EMR) issues reporting was developed in line with the workflow of the existing EMR reporting process. The helpdesk comprises four main components: user data, report data, problem solving data, and reporting data. Implications: the helpdesk application is necessary to ensure optimal coordination of users and the EMR manager, thereby minimizing the problems that occur. Conclusion: the designed helpdesk application successfully addresses EMR reporting issues that were previously handled manually.

In the digitalization era, transforming Hospital Information Systems presents challenges in managing physical medical records, including limited digital infrastructure, unprepared human resources, and difficulties integrating manual systems with digital technologies. These challenges pose risks to data accuracy, accessibility, and patient safety during the shift toward more efficient and secure digital systems. Purpose: This study aims to evaluate the storage and management system of physical medical records amid the digitalization process and identify barriers and potential integration with digital systems. Method: A qualitative approach was used through a case study in a regional public hospital. Data were collected via in-depth interviews, observations, and document analysis. Results: Findings indicate that hospitals still rely on physical medical records due to limited infrastructure, cybersecurity concerns, and insufficient staff training. About 65% of respondents identified patient data protection as a major issue, while limited training reduced the effectiveness of electronic system implementation. Implications: The findings emphasize the need to strengthen IT capacity and provide ongoing staff training, supported by budgets for infrastructure and skill development. Conclusion: Although digitalization is inevitable, physical medical records remain essential during the transition. Integration can be achieved through hybrid systems using technologies such as barcode scanning to enhance data accuracy, efficiency, and security.

Aims: To evaluate alterations in weight, body composition, and nutritional parameters, and to identify the clinical, anthropometric, and dietary determinants of these changes among patients with breast cancer (BC) treated with adjuvant chemotherapy (AC). Methods: This study involved a cohort of 252 BC patients treated with AC. Anthropometric measurements were collected using standardized equipment and body composition was evaluated via bioelectrical impedance analysis (BIA). Clinical data were retrieved from the hospital's information system. Adherence to the Mediterranean Diet (MedDiet) was evaluated using a validated food frequency questionnaire. Results: Post-AC, 29.3% and 22.7% of patients experienced moderate/high weight gain (WG ≥ 5%) and moderate/severe weight loss (MSWL ≥ 5%), respectively. The prevalence of patients with low muscle mass (LMM) and sarcopenic obesity increased from 11.5% and 2.8% pre-treatment to 15.1% and 7.9% post-treatment, respectively. Younger women (aged 24 – 39 years) exhibited reduced odds of experiencing MSWL (OR = 0.31, 95%CI: 0.08 – 1.20) and LMM (OR = 0.19, 95%CI: 0.02 – 1.61) compared to those aged ≥ 60 years. Patients with BC stages I and II were associated with increased odds of WG (OR = 2.10, 95%CI: 0.44 – 9.89; and OR = 2.13, 95%CI: 0.46 – 9.68, respectively). Relative to obese individuals, normal-weight/underweight individuals exhibited a significantly higher likelihood of WG (OR = 2.29; 95%CI: 1.03 – 5.08) and MSWL (OR = 3.23, 95%CI: 1.19 – 8.80), but a lower likelihood of LMM (OR = 0.33, 95%CI: 0.13 – 0.87). The Anthracycline-Taxane and Monoclonal-antibodies-Taxane regimens were associated with higher odds of MSWL (OR = 3.64, 95%CI: 0.35 – 36.98, and OR = 2.63, 95%CI: 1.03 – 6.72, respectively). Low and moderate adherence to the MedDiet were independently associated with an elevated risk of both WG and MSWL. Conclusions: A substantial proportion of patients with BC experience significant weight fluctuation or deterioration in muscle mass following AC. These adverse outcomes are modulated by patient age, cancer stage or duration, baseline BMI, the specific chemotherapy regimen employed, and MeDiet adherence. Keywords: Adjuvant chemotherapy; Body composition; Body Mass Index; Breast cancer; Mediterranean diet; Weight change.

BackgroundHypothermia is one of the most common postoperative complications in the post-anesthesia care unit (PACU), and preventive measures are crucial for improving patient outcomes. Anesthesia nurses play a vital role in the prevention and management of hypothermia, but the factors influencing their behavior have not yet been systematically explored.ObjectiveThis study aimed to explore the actual experiences of the implementation in hypothermia prevention measures by PACU nurses in post-anesthesia care unit, providing a basis for the formulation of management plans.MethodsA descriptive qualitative research design was employed, with nurses in the post-anesthesia care unit of a general hospital in Guizhou Province, China, selected as the study subjects. Data analysis was performed directed content analysis.ResultsThree major categories emerged, corresponding to the predisposing, reinforcing, and enabling factors outlined by the PRECEDE-PROCEED model. Predisposing factors included knowledge gaps on hypothermia, complexity of evidence-to-practice translation, conflicts between mindset and habitual practices. Enabling factors included insufficient human and equipment resources, lack of targeted and continuous training, lack of standardized guidelines for hypothermia prevention management processes, limitations in hospital information systems and lack of intelligent monitoring functions. Reinforcing factors included healthcare collaboration and peer support promote hypothermia prevention practices, lack of supervisory network for hypothermia prevention.ConclusionTo enhance hypothermia prevention practices in PACU, nursing administrators should implement need-based multidisciplinary training programs, optimize staffing, equipment resources, develop intelligent decision-support systems, and establish digitalized monitoring networks to support continuous quality improvement.

The growing complexity of healthcare information technology (IT) infrastructures, driven by the proliferation of electronic health records, connected medical devices, telemedicine platforms, and cloud-based clinical services, has created an urgent need for advanced analytical methods capable of modeling, optimizing, and safeguarding large-scale interconnected systems. Graph theory, with its ability to mathematically represent relationships among discrete components, provides a powerful foundation for understanding the structural and functional behavior of healthcare IT networks. This study examines how graph-theoretic principles and network analysis techniques, integrated with artificial intelligence (AI), can be systematically applied to the intelligent design and evaluation of healthcare IT infrastructures where scalability, reliability, data security, and performance efficiency are critical. The research models core healthcare infrastructure components—including clinical servers, hospital information systems, medical IoT devices, data centers, and communication links—as graph structures that capture both connectivity and operational dependencies. Network analysis metrics such as centrality measures, clustering coefficients, cut-sets, and shortest-path algorithms are applied to identify critical nodes, communication bottlenecks, and points of vulnerability that may compromise service continuity or patient safety. In parallel, spectral and flow-based graph analyses are employed to assess load distribution, latency patterns, and potential failure propagation across healthcare networks. Building upon these graph-derived insights, AI and machine learning techniques are incorporated to enable predictive maintenance, intelligent load balancing, and adaptive infrastructure planning. Dynamic graph modeling allows the system to capture temporal variations in healthcare data traffic, detect anomalies in real time, and anticipate structural weaknesses before they escalate into operational failures. The framework also supports comparative evaluation of alternative healthcare network architectures—including centralized, distributed, hybrid, and software-defined models—revealing critical trade-offs in responsiveness, fault tolerance, data availability, and scalability. The findings demonstrate that the integration of graph theory, network analysis, and AI significantly enhances the intelligence, resilience, and efficiency of healthcare IT infrastructure design. By providing a rigorous, quantifiable basis for architectural decision-making, this approach aligns technological capabilities with clinical performance objectives and regulatory requirements. The study concludes that graph-theoretic and AI-driven methodologies are essential components of next-generation healthcare IT engineering, enabling secure, adaptive, and future-ready digital health ecosystems.