Search Results (1,271)

Background and Objectives: Cesarean hysterectomy is a critical intervention often required to manage life-threatening postpartum hemorrhage (PPH) due to complications such as uterine atony, abnormal placental implantation, or traumatic rupture. Although lifesaving, the procedure is associated with significant risks and complications. This study investigates the causes and outcomes of cesarean hysterectomy, focusing on complications arising from the procedure. Materials and Methods: A retrospective analysis was conducted on 82 women who underwent cesarean hysterectomy at Afzali Pour Hospital between 2018 and 2022. All patients were followed for 42 days post-surgery to evaluate outcomes and complications. Data were extracted from electronic medical records, encompassing demographic, obstetric, and clinical details, including age, body mass index, previous cesarean sections, indications for cesarean deliveries, causes of hysterectomy, and complications. The primary outcome was to determine the causes of cesarean hysterectomy, while the secondary outcome assessed the complications associated with the procedure. Stepwise logistic regression analysis was utilized to identify significant predictors of complications. Results: The study included 82 women who underwent cesarean hysterectomy. The mean age of the participants was 35.2 years (SD = 5.4), with a range from 24 to 48 years. The average BMI was 29.1 kg/m² (SD = 4.3), with 45% of the women classified as overweight or obese (BMI ≥ 25). The majority of the patients (70%) had a history of two or more previous cesarean sections, and the most common indication for cesarean hysterectomy was abnormal placentation, including placenta accreta (58%). Uterine rupture was reported in 13% of the cases. In terms of complications, bladder injury was the most common, occurring in 33.33% of women, followed by fever (20%), ureteral injury (13.33%), and hematoma (8.89%). Stepwise logistic regression analysis revealed that higher BMI significantly increased the odds of the outcome (OR = 4.18, 95% CI: 1.66–10.51, p = 0.002), and the number of previous cesarean sections was also a significant predictor (OR = 2.30, 95% CI: 1.17–4.53, p = 0.016). Conclusions: Placenta accreta and previa were the most frequent causes of cesarean hysterectomy, with bladder injury and fever being the most common complications. A higher number of previous cesareans and higher BMI significantly increase the likelihood of complications. Understanding these risk factors can improve patient management and surgical outcomes, highlighting the importance of careful monitoring and preoperative planning in women with a history of cesarean deliveries. Full article

We sought to incorporate a community-based solution with a family health history (FHH) clinical support program (MeTree) integrated into well-patient appointments with the novel partnership of a public health state-level health information exchange (HIE). The Arkansas—Making History pilot project tested informatics compatibility among these systems and the patients’ electronic medical record (EPIC) in a rural clinic in the north central region of the state, having the state HIE as a means for patients to store and share their FHHs across multiple healthcare providers with updates in real time. We monitored for unexpected issues during the pilot and asked for the perspectives of patients and healthcare providers throughout the project to have a clear understanding of how to implement this project on a larger scale. The greatest barrier to project implementation was the inability of the state HIE to host or share the FHH data. We compensated for the lack of systems compatibility and documented valuable information about patient acceptability and usability of the MeTree platform, as well as gleaning important clinical outcome data from those who completed MeTree FHH accounts in an underserved area. Rural patients need additional technological support in the larger scaling of this project, both in available linkages to community clinics with patient-controlled options for how their data is stored and shared and in Internet connectivity and software options available for ease of use. Full article

Greater than 650 million individuals worldwide are categorized as obese, which is associated with significant health, economic, and social challenges. Given its overlap with leading comorbidities such as heart disease, innovative solutions are necessary to improve risk prediction and management strategies. In recent years, artificial intelligence (AI) and machine learning (ML) have emerged as powerful tools in healthcare, offering novel approaches to chronic disease prevention. This narrative review explores the role of AI/ML in obesity risk prediction and management, with a special focus on childhood obesity. We begin by examining the multifactorial nature of obesity, including genetic, behavioral, and environmental factors, and the limitations of traditional approaches to predict and treat morbidity associated obesity. Next, we analyze AI/ML techniques commonly used to predict obesity risk, particularly in minimizing childhood obesity risk. We shift to the application of AI/ML in obesity management, comparing perspectives from healthcare providers versus patients. From the provider’s perspective, AI/ML tools offer real-time data from electronic medical records, wearables, and health apps to stratify patient risk, customize treatment plans, and enhance clinical decision making. From the patient’s perspective, AI/ML-driven interventions offer personalized coaching and improve long-term engagement in health management. Finally, we address key limitations and challenges, such as the role of social determinants of health, in embracing the role of AI/ML in obesity management, while offering our recommendations based on our literature review. Full article

Background: Pressure injuries (PIs) are increasing worldwide, and there has been no significant improvement in preventing them. Traditional assessment tools are widely used to identify a patient at risk of developing a PI. This study aims to construct a novel fused multi-channel prediction model of PIs in adult hospitalized patients using machine learning algorithms (MLAs). Methods: A multi-phase quantitative approach involving a case–control experimental design was used. A first-hand dataset was collected retrospectively between March/2022 and August/2023 from the electronic medical records of three hospitals in Palestine. Results: The total number of patients was 49,500. A balanced dataset was utilized with a total number of 1110 patients (80% training and 20% testing). The models that were developed utilized eight MLAs, including linear regression and support vector regression (SVR), logistic regression (LR), random forest (RF), gradient boosting (GB), K-nearest neighbor (KNN), decision tree (DT), and extreme gradient boosting (XG boosting) and validated with five-fold cross-validation techniques. The best model was RF, for which the accuracy was 0.962, precision was 0.942, recall was 0.922, F1 was 0.931, area under curve (AUC) was 0.922, false positive rate (FPR) was 0.155, and true positive rate (TPR) was 0.782. Conclusions: The predictive factors were age, moisture, activity, length of stay (LOS), systolic blood pressure (BP), and albumin. A novel fused multi-channel prediction model of pressure injury was developed from different datasets. Full article

Background and Objectives: The prevalence of and risk factors for immune checkpoint inhibitor-associated diarrhea and colitis (IMDC) in the Chinese population are unclear. This study aimed to estimate IMDC incidence and identify potential risk factors. Materials and Methods: We reviewed the electronic medical records from Beijing Friendship Hospital (2015–2022) to identify the patients treated with immune checkpoint inhibitors. The primary outcome was IMDC occurrence. The demographics, cancer type, baseline labs, and concurrent medications were analyzed. The univariable and multivariable analyses validated the associated factors. Results: Among 1186 patients (median follow-up: 217 days), the IMDC incidence was 4.6%, with colitis at 0.67%. Digestive system tumors increased the IMDC risk (OR 2.79, 95% CI 1.42–5.75, p = 0.004), while platinum agents decreased it (OR 0.41, 95% CI 0.21–0.78, p = 0.008). PPIs, antibiotics, NSAIDs, and glucocorticoids showed no significant association. Colitis was the third most common irAE, leading to ICI discontinuation (15.6%). Conclusions: IMDC prevalence is 4.6% in the Chinese population, the third most frequent irAE causing ICI discontinuation. Digestive tumors and platinum agents are risk and protective factors, respectively, while other medications show no significant impact. Full article

Background/Objectives: During the early phase of the coronavirus disease 2019 (COVID-19) pandemic, people were advised to stay at home and the American College of Surgeons suggested the nonoperative management (NOM) of uncomplicated appendicitis. We hypothesized that patients presented with more cases of complicated appendicitis during the early phase of COVID-19 compared with the previous year; we further hypothesized that more patients had NOM. Methods: Adults diagnosed with appendicitis were retrospectively reviewed from electronic medical records throughout a single county-wide hospital system. The pre-pandemic period (3 January 2019–30 June 2019, PRE) was compared with the pandemic period (3 January 2020–30 June 2020, POST). The primary outcome was AAST grade of appendicitis. Results: There were 278 cases of appendicitis in PRE and 269 in POST. The rate of complicated appendicitis (grades II–V) was higher in POST (39% vs. 30%, p = 0.0375), most prominently in the northern hospitals in the county (41% vs. 27%, p = 0.004), with non-operative management in six (3.2%) cases. Grades III–V, consistent with perforation, were seen in 33% of POST vs. 27% of PRE cases (p = 0.098). Grade I appendicitis was managed non-operatively in only six (1.6%) patients. There were fewer readmissions in POST (4% vs. 8%, p = 0.0427) and no mortalities during the study period. Conclusions: There was a significant increase in presentation with complicated appendicitis during the early phase of the COVID-19 pandemic in the northern hospitals in the county. There was no increase in NOM of uncomplicated appendicitis and no change in hospital LOS but there were fewer readmissions during COVID-19. Full article

Background: During the 2023–2024 season, the influenza epidemic in South Korea peaked earlier, and the influenza vaccination rate among individuals aged ≥ 65 was high (82.2%). However, data on real-world vaccine effectiveness against influenza are lacking. Methods: From November 2023 to April 2024, we conducted a multicenter retrospective case–control study on adult patients aged ≥ 18 years who presented with influenza-like illness at seven medical centers as a part of a hospital-based influenza morbidity and mortality surveillance (HIMM) program in South Korea. Demographic and clinical data were collected from questionnaire surveys and electronic medical records. Using a test-negative design, we assessed the effectiveness of the 2023–2024 seasonal influenza vaccine, with age, sex, and comorbidities included as covariates. Results: A total of 3390 participants were enrolled through the HIMM system, including 1695 patients with either rapid antigen test (RAT) or real-time reverse-transcription polymerase chain reaction (RT-PCR) positive results and controls matched for age, sex, and months of registration. Among the 1696 influenza-positive patients, 1584 (93.5%) underwent RAT, with 88.9% testing positive for influenza A and 11.1% for influenza B. During the study periods, the overall vaccine effectiveness (VE) was 24.3% (95% confidence interval (CI), 11.5 to 35.2). The VE was insignificant when limited to older adults aged ≥ 65 years (13.5%; 95% CI, −17.9 to 36.6). In the subgroup analysis by subtype, the VE was 19.0% (95% CI, 5.0 to 31.0) for influenza A and 56.3% (95% CI, 35.3 to 70.6) for influenza B. Notably, influenza VE was 20.4% (95% CI, 2.9 to 34.8) in the early period (November to December) but decreased to 12.4% (95% CI, −14.9 to 33.2) in the late period (January to April). Conclusion: During the 2023–2024 season, the influenza vaccine showed a modest effectiveness (24.3%) against laboratory-confirmed influenza, which was particularly higher for influenza B. Because the VE was insignificant in older adults, particularly during the late period, better immunogenic influenza vaccines with longer-lasting protection should be considered. Full article

Background/Objectives: Chimeric antigen receptor T-cell therapy (CAR-T) has become a key treatment option for relapsed/refractory multiple myeloma (RRMM), but factors impairing T-cell fitness may diminish efficacy. Our exploratory analysis aimed to evaluate the impact of prior treatment with a selinexor-containing regimen on CAR-T outcomes for RRMM patients. Methods: Data for this retrospective cohort study were sourced from electronic medical records at two US academic centers. Kaplan–Meier estimates assessed duration of response (DOR), progression-free survival (PFS), and overall survival (OS), reported as medians with interquartile ranges (IQRs). Cox proportional hazards regression analyzed factors potentially associated with PFS and OS, reported as hazard ratios (HRs) with 95% confidence intervals (CIs). Results: Among 45 patients exposed to selinexor before undergoing BCMA-directed CAR-T, median therapy line numbers for selinexor use and CAR-T were 7 and 9, respectively, with 24.4% receiving selinexor as part of bridging. At median follow-up of 68 months, median PFS and OS post CAR-T were 8.0 (IQR 3.1–39.5) and 35.9 (IQR 14.2–NR) months, respectively. Overall response rate to CAR-T was 89%, with a median DOR of 8.1 months (IQR 2.9–39.0). In our multivariable model, patients who received a selinexor-based regimen in the line of therapy preceding CAR-T showed a trend toward reduced risk of death (HR = 0.08; 95% CI 0.02–0.46) and/or disease progression (HR = 0.40; 95% CI 0.14–1.09). Conclusions: Prior selinexor exposure does not appear to compromise CAR-T outcomes in heavily pretreated RRMM, suggesting potential T-cell sparing. Our findings warrant larger, prospective studies to determine whether preemptive selinexor treatment can optimize CAR-T efficacy. Full article

Recommending safe and effective medication (drug) combinations is a key application of artificial intelligence in healthcare. Current methods often focus solely on recommendation accuracy while neglecting drug–drug interactions (DDIs), or overly prioritize reducing DDIs at the cost of model performance. Therefore, we propose the Attention-guided Multi-Graph collaborative decision Network (AMGNet) for safe medication recommendation, which strikes a balance between improving recommendation accuracy and minimizing DDIs. Specifically, AMGNet designs a patient feature encoder that utilizes a transformer encoder–decoder architecture to learn the temporal dependencies of longitudinal medical features from patient visits, effectively capturing the patient’s medication history and health status to enhance recommendation accuracy. AMGNet is also equipped with a medication feature encoder that integrates diverse knowledge graphs of drug molecular structure, electronic health records(EHRs), and drug–drug interactions(DDIs) through multi-graph representation learning and contrastive learning methods, further reducing the DDI rate of the recommended medication combinations and mitigating the risks associated with drug co-administration. We conducted extensive experiments on the widely used MIMIC-III and MIMIC-IV clinical medical datasets. The results demonstrate that AMGNet achieves competitive performance. Additionally, ablation studies and detailed case analyses further confirm that AMGNet offers high precision and safety in medication recommendation. Full article

Background: Sickle cell disease (SCD) is a rare autosomal recessive disorder that is common in countries with consanguineous marriages. It leads to various complications, including painful episodes, infections, delayed growth, stroke, and organ damage, which contribute to high healthcare utilization and costs. In Saudi Arabia, the prevalence of SCD is notably high, largely due to the frequency of consanguineous marriages. However, there has not yet been a study estimating the direct medical costs of managing SCD based on real-world data. This study aims to assess these costs in Saudi Arabia. Methods: Data were collected from electronic medical records (EMRs) at a university-affiliated tertiary care center. A micro-costing approach was used to estimate the direct medical costs (e.g., laboratory tests, imaging, emergency department visits, hospitalizations, prescription medications, outpatient visits, etc.) retrospectively over a 12-month follow-up period. The baseline characteristics of the patients were presented using frequencies and percentages. The costs of different healthcare services were analyzed using means and the 95% confidence intervals. A generalized linear model (GLM) with a gamma distribution was utilized to examine the association between the overall costs and patient characteristics (e.g., age, gender, duration of illness, surgeries, blood transfusions, etc.), allowing for the estimation of the adjusted mean costs. Results: A total of 100 patients met the inclusion criteria and were included in the analysis. The mean age of the patients was 10.21 years (±6.87 years); 53% were male, and a substantial majority (96%) had the HbSS genotype. Sixty-one percent of the patients had undergone at least one red blood cell (RBC) exchange transfusion, while 21% had undergone surgical procedures, including tonsillectomy, splenectomy, and cholecystectomy. Additionally, 45% had experienced at least one vaso-occlusive crisis (VOC), and 59% had been hospitalized at least once in the past 12 months. Factors such as the frequency of laboratory tests and imaging studies, the length of hospital stay (LOS), the rate of emergency department (ED) visits, surgical procedures, the number of prescription medications, and the frequency of blood transfusions were all significant predictors of higher direct medical costs (p < 0.05). The estimated mean annual direct medical costs per patient were USD 26,626.45 (95% CI: USD 22,716.89–USD 30,536.00). After adjusting for various factors, including age, gender, duration of illness, frequency of lab and imaging tests, LOS, ED visits, surgical procedures, number of prescription medications, rates of VOCs, and RBC exchange transfusions, the adjusted mean annual direct medical cost per patient was calculated to be USD 14,604.72 (95% CI: USD 10,943.49–USD 19,525.96). Conclusions: The results of this study emphasize the substantial direct medical costs linked to sickle cell disease (SCD), which are greatly affected by the frequency of related complications. These insights should motivate policymakers and healthcare researchers to assess both the national direct and indirect costs associated with SCD, especially given the significant number of SCD patients in Saudi Arabia. Full article

Approximately 50% of Alzheimer’s disease (AD) patients develop psychotic symptoms, leading to a subtype known as psychosis in AD (AD + P), which is associated with accelerated cognitive decline compared to AD without psychosis. Currently, no FDA-approved medication specifically addresses AD + P. This study aims to improve psychosis predictions and identify potential therapeutic agents using the DeepBiomarker deep learning model by incorporating drug–target interactions. Electronic health records from the University of Pittsburgh Medical Center were analyzed to predict psychosis within three months of AD diagnosis. AD + P patients were classified as those with either a formal psychosis diagnosis or antipsychotic prescriptions post-AD diagnosis. Two approaches were employed as follows: (1) a drug-focused method using individual medications and (2) a target-focused method pooling medications by shared targets. The updated DeepBiomarker model achieved an area under the receiver operating curve (AUROC) above 0.90 for psychosis prediction. A drug-focused analysis identified gabapentin, amlodipine, levothyroxine, and others as potentially beneficial. A target-focused analysis highlighted significant proteins, including integrins, calcium channels, and tyrosine hydroxylase, confirming several medications linked to these targets. Integrating drug–target information into predictive models improves the identification of medications for AD + P risk reduction, offering a promising strategy for therapeutic development. Full article

Background/Objectives: The standard evaluation of children and adolescents suspected of having thyroid disorders consists of anthropometric measurements. Body composition features provide additional information for enhanced therapeutic management. We explored the muscle-to-fat ratio of pediatric patients referred for thyroid disorders and its interaction with thyroid function and metabolic syndrome components. Methods: This retrospective cross-sectional study consisted of 147 pediatric subjects (ages 5–19 years) diagnosed with childhood-onset thyroid disorders treated at a tertiary medical center. Sociodemographic, clinical and laboratory data [thyroid-stimulating hormone (TSH), free T4 (FT4), and lipid profile] were extracted from the electronic medical records. Body composition was measured using bioimpedance analysis (Tanita MC-780 MA and GMON Professional Software). Body mass index (BMI), appendicular muscle mass (ASMM), and muscle-to-fat ratio (MFR) were converted to z-scores. Results: The diagnoses included Hashimoto thyroiditis (30.6%), subclinical hypothyroidism (26.5%), congenital hypothyroidism (21.7%), and Graves’ disease (21%). Based on BMI z-scores, 31.3% of the cohort was overweight or obese. The TSH levels were positively correlated with the BMI z-scores (r = 0.238, p = 0.005) and negatively with the MFR z-scores (r = 0.215, p = 0.012). The ASMM z-scores were negatively associated with the FT4 levels (r = −0.255, p = 0.003). Dyslipidemia was prevalent. TSH was correlated with LDL cholesterol (r = 0.472, p < 0.001) and triglycerides (r = 0.232, p = 0.05). Conclusions: Elevated thyroid-stimulating levels were linked to higher BMI and lower MFR levels. Our findings on the relationship between thyroid function and lipid profile underscore the necessity of optimizing thyroid balance and implementing targeted lifestyle interventions to improve body composition in young patients with thyroid disorders. Full article

Traditional Chinese medicine (TCM) gathers patient information through inspection, olfaction, inquiry, and palpation, analyzing and interpreting the data to make a diagnosis and offer appropriate treatment. Traditionally, the interpretation of this information relies heavily on the physician’s personal knowledge and experience. However, diagnostic outcomes can vary depending on the physician’s clinical experience and subjective judgment. This study employs AI methods to focus on localized tongue assessment, developing an automatic tongue body segmentation using the deep learning network “U-Net” through a series of optimization processes applied to tongue surface images. Furthermore, “ResNet34” is utilized for the identification of “cold”, “neutral”, and “hot” constitutions, creating a system that enhances the consistency and reliability of diagnostic results related to the tongue. The final results demonstrate that the AI interpretation accuracy of this system reaches the diagnostic level of junior TCM practitioners (those who have passed the TCM practitioner assessment with ≤5 years of experience). The framework and findings of this study can serve as (1) a foundational step for the future integration of pulse information and electronic medical records, (2) a tool for personalized preventive medicine, and (3) a training resource for TCM students learning to diagnose tongue constitutions such as “cold”, “neutral”, and “hot”. Full article

Background/Objectives: Limited evidence is available regarding insulin total daily dose (TDD), or the factors associated with TDD, among adults with type 2 diabetes (T2D) using multiple daily injections of insulin (MDI). Our aim was to determine the percentage of adults in the United States (US) with T2D who are prescribed MDI, their prescribed insulin TDD, and potential factors associated with TDD. Methods: This retrospective cohort study used deidentified data from the US IQVIA ambulatory electronic medical record database to study adults (≥18 years) with T2D initiating MDI (≥3 daily basal-plus-prandial insulin injections) from 1 January 2017 to 1 July 2022. The TDD was calculated from first evidence of MDI (index date). We used a generalized linear model regression analysis to model the relationship between TDD and clinically relevant factors associated with TDD. Results: During the study period, of 3,339,663 adults with T2D, 451,769 (13.5%) had ≥1 basal insulin prescriptions, 206,000 (6.2%) had both basal and prandial insulin prescriptions, and 41,215 (1.2%) were prescribed MDI (mean age, 58 years; 52% women; 62% White/Caucasian, 14% African American; mean body mass index [BMI], 34 kg/m²). Mean TDD was 96 units (1.0 units/kg/day); median TDD was 80 units (interquartile range, 54–124). In the regression analysis (model R², 0.14), factors predicting lower TDD included female sex, African American race, and prior 6-month (pre-index) prescriptions of sulfonylurea, metformin, or 2–3 noninsulin glucose-lowering medications. Predictors of greater TDD included increasing BMI, age 30–64 years, and pre-index SGLT2 inhibitor or GLP-1 RA prescription. Conclusions: Among US adults with T2D, 1.2% were prescribed MDI, with a wide range of TDD and median TDD of 80 units. Further research in other populations and using other data sources is warranted to explore prescribed insulin TDD for T2D and to examine other potentially relevant predictors of TDD. Full article

Background: Hospital-acquired pressure injuries (HAPIs) affect approximately 2.5 million patients annually in the United States, leading to increased morbidity and healthcare costs. Current rule-based screening tools, such as the Braden Scale, lack sensitivity, highlighting the need for improved risk prediction methods. Methods: We developed and externally validated a machine learning model to predict HAPI risk using longitudinal electronic medical record (EMR) data. This study included adult inpatients (2018–2023) across five hospitals within a large health system. An automated pipeline was built for EMR data curation, labeling, and integration. The model employed XGBoost with recursive feature elimination to identify 35 optimal clinical variables and utilized time-series analysis for dynamic risk prediction. Results: Internal validation and multi-center external validation on 5510 hospitalizations demonstrated AUROC values of 0.83–0.85. The model outperformed the Braden Scale in sensitivity and F1-score and showed superior performance compared to previous predictive models. Conclusions: This is the first externally validated, cross-institutional HAPI prediction model using longitudinal EMR data and automated pipelines. The model demonstrates strong generalizability, scalability, and real-time applicability, offering a novel bioengineering approach to improve HAPI prevention, patient care, and clinical operations. Full article