Patient-specific Risk Research Articles

Development and Validation of a Multimodal Machine Learning Model for Diagnosing and Assessing Risk of Crohn's Disease in Patients With Perianal Fistula.

Diagnosis of Crohn's disease (CD) can pose challenges, particularly when perianal fistula is the initial presentation. To develop and validate a predictive model, establishing a visual web tool for early diagnosis of CD in patients presenting with perianal fistula. This retrospective, multicentre validation study involved patients diagnosed with either perianal fistulising CD or cryptoglandular fistula who underwent initial perianal fistula surgery subsequent to rectal MRI at three Chinese centres from September 2016 to December 2020. A random forest classification model was trained on the derivation cohort (n = 550), randomly split into training and test sets at a 7:3 ratio. Validation utilised data from two external centres (n = 300). Model interpretation employed the Shapley Addictive explanation (SHAP) framework. The validated model was integrated into a web tool for calculating patient-specific risk. In the derivation cohort, SHAP analysis highlighted rectal wall ulceration, rectal wall thickening, submucosal fistula, and T2 hyperintensity as risk factors, while age was identified as protective. A random forest classification model developed using these top 5 features achieved an AUROC of 0.9425 (95% CI: 0.8943-0.9906). In the validation cohort, the model performed well with AUROC values of 0.9187 (95% CI: 0.8620-0.9754) and 0.9341 (95% CI: 0.8876-0.9806), respectively. We developed a publicly accessible web-based application. We have developed a multimodal machine learning model and a web tool that can predict and present CD risk in patients initially presenting with perianal fistula.

Multicentric national Italian analysis of textbook outcome in colorectal cancer surgery: The ATOCCS Study protocol on behalf of the Italian Surgical Association (ACOI, Associazione Chirurghi Ospedalieri Italiani)

Background: Colorectal cancer (CRC) is a leading cause of cancer-related mortality worldwide. Despite advancements in surgical techniques and perioperative care, patient outcomes vary significantly across healthcare institutions. Textbook outcome (TO), a composite metric representing an ideal postoperative course, has been proposed as a standard for assessing the quality of surgical care in CRC. However, its impact on patient outcomes remains underexplored, particularly across diverse clinical settings. Objective: This study aims to explore surgeons’ adherence to TO criteria and identify patient-specific and procedural risk factors associated with TO failure. The study will also evaluate TO as a quality metric in CRC surgery, examining its association with short-term and long-term clinical outcomes in a multicenter cohort. Methods: This multicenter, retrospective cohort study will include all adult patients undergoing CRC resection between 1 January 2022 and 31 December 2024 across multiple centers in Italy, on behalf of the Italian Surgical Association (ACOI, Associazione Chirurghi Ospedalieri Italiani). The study will collect and analyze demographic, clinical, and surgical data to determine TO incidence and its association with key outcomes, including radical resection, 30-day mortality, morbidity, no reintervention, no ostomy placement, and a hospital stay of 14 days or less. Conclusions: This study will offer valuable insights into the utility of TO as a metric for evaluating the quality of care in CRC surgery. These findings may inform future guidelines and policies aimed at improving CRC surgical outcomes.

Comparing Percutaneous Coronary Intervention and Coronary Artery Bypass Graft in Treatment of Non-ST-segment Elevation Acute Coronary Syndromes: A Systematic Review and Meta-Analysis Study.

The objective of this study is to compare the effectiveness and safety of percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) in the treatment of non-ST-segment elevation acute coronary syndromes (NSTE-ACS). A literature search was conducted across PubMed, Scopus, and Web of Science, covering studies up to June 2024. Studies comparing PCI and CABG in patients with NSTE-ACS were included, focusing on clinical outcomes such as mortality, myocardial infarction (MI), cerebrovascular accidents (CVA), and the need for repeat revascularization. Data extraction and quality assessment were performed. Statistical analysis was conducted using R software, with the Mantel-Haenszel method and random-effects model employed to pool effect sizes and assess heterogeneity. A total of 15 studies met the eligibility criteria, including 48,891 patients. The pooled risk ratio (RR) for mortality showed no significant difference between PCI and CABG (RR = 1.09, 95% CI: 0.90-1.19, p = 0.28). CABG was associated with a significantly lower risk of subsequent MI (RR = 0.56, 95% CI: 0.38-0.61, p < 0.01) and the need for repeat revascularization (RR = 2.94, 95% CI: 2.30-3.76, p < 0.01). Conversely, PCI had a lower associated risk of CVA (RR = 0.58, 95% CI: 0.42-0.79, p < 0.01). High heterogeneity was observed in mortality outcomes, indicating variability among studies. The findings suggest that while PCI and CABG have comparable mortality risks in NSTE-ACS patients, CABG offers superior protection against myocardial infarction and the need for repeat revascularization, whereas PCI is associated with a lower risk of cerebrovascular accidents. These results underscore the importance of individualized patient assessment in choosing the optimal revascularization strategy, considering patient-specific risk factors and clinical profiles.

Evaluating diclofenac's risks in COVID-19: strategies for mitigating adverse outcomes.

The use of nonsteroidal anti-inflammatory drugs (NSAIDs), particularly diclofenac, during the COVID-19 pandemic has raised concerns due to its potential to worsen disease progression. This commentary evaluates key risks associated with diclofenac and highlights the critical role of pharmacists in mitigating adverse outcomes through careful medication management and patient education. Diclofenac poses unique risks due to its ability to generate reactive oxygen species (ROS), leading to oxidative stress and mitochondrial dysfunction. In COVID-19, a disease characterized by hyperinflammation, these effects may exacerbate systemic inflammation, contributing to severe outcomes. Moreover, diclofenac's known association with increased cardiovascular risks, such as myocardial infarction and stroke, is especially concerning in patients with COVID-19, who are predisposed to thrombotic complications. The drug's hepatotoxic potential adds another layer of concern, particularly in patients with pre-existing liver dysfunction or those at higher risk due to COVID-19-related liver involvement. Pharmacists play a pivotal role in addressing these risks by conducting thorough medication reviews and assessing patient-specific risk factors. They can guide clinicians and patients toward safer alternatives, such as ibuprofen or naproxen, which demonstrate a lower oxidative and cardiovascular burden. Patient education is equally critical; pharmacists should counsel individuals on potential adverse effects of diclofenac, such as cardiovascular symptoms, renal dysfunction, and liver complications, while advising on lifestyle modifications and adjunctive therapies to reduce NSAID dependence. Additionally, pharmacists contribute to pharmacovigilance by monitoring patients for adverse drug reactions and reporting safety concerns to improve NSAID usage guidelines during the pandemic. By adopting a personalized approach to NSAID therapy, pharmacists can minimize risks and enhance patient safety, ultimately improving outcomes in the management of COVID-19 and other inflammatory conditions. This underscores their indispensable role in optimizing care during complex clinical scenarios.

Routine 36‐week scan: prediction of small‐for‐gestational‐ age neonate

Routine 36‐week scan: prediction of small‐for‐gestational‐ age neonate

Pneumonitis after normofractionated radioimmunotherapy: a method for dosimetric evaluation

BackgroundPost-Therapy-Pneumonitis (PTP) is a critical side effect of both, thoracic radio(chemo)therapy (R(C)T) and immune checkpoint inhibition (ICI). However, disease characteristics and patient-specific risk factors of PTP after combined R(C)T + ICI are less understood. Given that RT-triggered PTP is strongly dependent on the volume and dose of RT [1], driven by inflammatory mechanisms, we hypothesize that combination therapy of R(C)T with ICI influences the dose-volume-effect correlation for PTP. This study focuses on the development of a method for evaluation of alterations of dosimetric parameters for PTP after R(C)T with and without ICI.Methods and materialsPTP volumes were delineated on the follow-up diagnostic Computed Tomography (CT) and deformably matched to the planning CT for patients with PTP after thoracic R(C)T + ICI or R(C)T. Dose data was converted to 2-Gy equivalent doses (EQD2) and dosimetrically analyzed. Dosimetric and volumetric parameters of the segmented PTP volumes were analyzed. The method was exemplarily tested on an internal patient cohort including 90 patients having received thoracic R(C)T + ICI (39) and R(C)T (51). Thirtytwo patients with PTP were identified for further analysis. Additional data on previous chemotherapy, RT, smoking status and pulmonary co-morbidity were conducted. A matched pair analysis with regard to planning target volumes (PTV) was conducted for curative intended (definitive) and palliative patient cohorts individually.ResultsThe presented method was able to quantify and compare the dosimetric parameters of PTP for the different therapies. For our study group, no significant differences between R(C)T + ICI and R(C)T only was observed. However, the dosimetric analysis revealed large volumetric fractions (55%) of the PTP volumes to be located outside of high dose (EQD2 < 40 Gy) regions for R(C)T + ICI. There was a non-significant trend towards increased area under the curve of the dose volume histogram (AUC) values for R(C)T + ICI compared to R(C)T only (3743.6 Gy∙% vs. 2848.8 Gy∙%; p-value = 0.171). In contrast to the data for the palliative intended treatment group, for definitive R(C)T + ICI, data tended towards increased volumes with higher doses.ConclusionsThe proposed method was capable to quantify dosimetric differences in the dose-volume-effect relationship of PTP for patients with R(C)T + ICI and patients with R(C)T only. In this exploratory analysis, no significant dosimetric differences within PTP volumes for the different groups could be observed. However, our observations suggest, that for safe application of thoracic R(C)T + ICI, further careful investigation of dosimetric prescription and analysis concepts with larger and conformer study groups is recommendable.

Abstract Su906: Patient and Arrest Characteristics Associated with Rearrest and Mortality Following Out of Hospital Cardiac Arrest

Background: Cardiac rearrest post-return of spontaneous circulation (ROSC) is a critical concern in emergency care, as it is associated with worse outcomes. Patient characteristics and arrest factors associated with rearrest remain poorly understood, making predicting rearrest after out of hospital cardiac arrest (OHCA) challenging. Aim: To examine rearrest patterns, identify patient-specific risk factors for rearrest, and assess the impact of rearrest on patient outcomes. Hypothesis: Patients that experience rearrest after OHCA will have greater cardiac comorbidities, unfavorable arrest characteristics, and greater mortality. Methods: We performed an observational single-site clinical trial of consecutive adult EMS OHCA patients with ROSC (1/01/2018 to 3/31/24). EMS reports, continuous EMS ECG recordings, and electronic health records were analyzed to determine rearrest occurrence at any time during EMS and ED care. Patient demographics, comorbidities, and arrest/rearrest characteristics were determined. The primary outcome was survival at 6 months. Statistical significance was determined using Chi-square and logistic regression (LR) analyses. Results: We examined 344 patients with OHCA and ROSC, of whom 173 patients rearrested and 171 did not. Patients experiencing rearrest were older (p&lt;0.003) and more likely to be female (p&lt;0.013). They were also more likely to have ED resuscitation medications and interventions (epinephrine, amiodarone, lidocaine, defibrillation, all p&lt;0.001) and a higher incidence of underlying cardiovascular comorbidities, including HTN and CAD (p&lt;0.03). Notably, PEA and asystole were the most common rearrest rhythms, and time from arrest to ROSC was greater (30±16 vs. 25±15 min, p&lt;0.007) in rearrest patients vs. those who did not rearrest. Patients with rearrest were less likely to receive TTM (p&lt;0.003) and importantly, had higher rates of in-hospital and six-month mortality (both p&lt;0.001). In a LR model, no rearrest (OR 2.4, 1.1-5.2), shockable arrest rhythm (OR 9.3, 4-22), female sex (OR 3.3, 1.6-6.5), and therapeutic hypothermia (OR 3.3, 1.7-6.4), were associated with survival (all p&lt;0.04). Conclusions: Rearrest is independently associated with mortality. Increased age, male sex, and underlying cardiac comorbidities were associated with rearrest. Patient and arrest characteristics may predict rearrest. Improved understanding of why these factors promote rearrest will be important to prevent rearrest to enhance OHCA outcomes.

Patient-specific multifactorial mortality risk assessment using classification and regression tree analysis in the context of ambulatory blood pressure monitoring.

Ambulatory blood pressure monitoring is essential for understanding blood pressure patterns beyond clinical visits, aiding in risk assessment, treatment evaluation, and managing hypertension. This retrospective cohort study aimed to identify risk factors for all-cause mortality and major cardiovascular events in patients who underwent ambulatory blood pressure monitoring. Eligible participants aged 18 or older, with an estimated glomerular filtration rate (eGFR) > 60ml/min/1.73m2, who underwent ambulatory blood pressure monitoring for various reasons, were included in the study. Data were gathered through telephone interviews, electronic health records, and the national health record system. Descriptive analysis and classification and regression tree modeling were used to uncover significant risk factors related to all-cause mortality and cardiovascular events, and to assess the model's performance compared to traditional Cox survival analysis. The study included 1291 patients, primarily male (51.8%) with a mean age of 61.1 ± 15.2years. During a mean follow-up of 46.9months, 76 (5.9%) patients died of any cause, and 195 (15.1%) had a cardiovascular event. The highest survival rates were observed in patients with a diastolic blood pressure (BP) dipping percentage between -2% and 29%, nighttime systolic BP variability below 32mmHg, and age below 72. Conversely, smokers with a diastolic BP dipping percentage below -10% showed the lowest survival rates. The best cardiovascular outcomes were observed in patients with diastolic BP dipping above -11%, nighttime mean systolic BP < 144mmHg, no statin use, normotensive status, and daytime mean heart rate ≥ 60bpm. Conversely, the worst outcomes were seen in patients with diastolic BP dipping below -11% and a morning surge ≥ 14mmHg. In all-cause mortality and cardiovascular event analysis, the combined model demonstrated excellent calibration and predictive power, like the classification and regression tree model and traditional analysis. These findings highlight the potential of a combined model for assessing mortality and cardiovascular event risk in patients who have undergone ambulatory blood pressure monitoring.

Multimodal AI/ML for discovering novel biomarkers and predicting disease using multi-omics profiles of patients with cardiovascular diseases

Cardiovascular diseases (CVDs) are complex, multifactorial conditions that require personalized assessment and treatment. Advancements in multi-omics technologies, namely RNA sequencing and whole-genome sequencing, have provided translational researchers with a comprehensive view of the human genome. The efficient synthesis and analysis of this data through integrated approach that characterizes genetic variants alongside expression patterns linked to emerging phenotypes, can reveal novel biomarkers and enable the segmentation of patient populations based on personalized risk factors. In this study, we present a cutting-edge methodology rooted in the integration of traditional bioinformatics, classical statistics, and multimodal machine learning techniques. Our approach has the potential to uncover the intricate mechanisms underlying CVD, enabling patient-specific risk and response profiling. We sourced transcriptomic expression data and single nucleotide polymorphisms (SNPs) from both CVD patients and healthy controls. By integrating these multi-omics datasets with clinical demographic information, we generated patient-specific profiles. Utilizing a robust feature selection approach, we identified a signature of 27 transcriptomic features and SNPs that are effective predictors of CVD. Differential expression analysis, combined with minimum redundancy maximum relevance feature selection, highlighted biomarkers that explain the disease phenotype. This approach prioritizes both biological relevance and efficiency in machine learning. We employed Combination Annotation Dependent Depletion scores and allele frequencies to identify variants with pathogenic characteristics in CVD patients. Classification models trained on this signature demonstrated high-accuracy predictions for CVD. The best performing of these models was an XGBoost classifier optimized via Bayesian hyperparameter tuning, which was able to correctly classify all patients in our test dataset. Using SHapley Additive exPlanations, we created risk assessments for patients, offering further contextualization of these predictions in a clinical setting. Across the cohort, RPL36AP37 and HBA1 were scored as the most important biomarkers for predicting CVDs. A comprehensive literature review revealed that a substantial portion of the diagnostic biomarkers identified have previously been associated with CVD. The framework we propose in this study is unbiased and generalizable to other diseases and disorders.

Machine learning based finite element analysis (FEA) surrogate for hip fracture risk assessment and visualization

Classical computational biomechanical approaches are costly and require a high level of expertise, often limiting their clinical application. A data-driven Machine Learning (ML) framework can serve as an effective alternative for disease diagnosis and prediction. This study aimed to develop an ML-based modeling approach integrating Quantitative Computed Tomography-based Finite Element Analysis (QCT-based FEA) to assess fall-induced hip fracture probability and probable facture locations in an elderly individual. Specifically, this study focused on predicting patient-specific hip fracture risk using fracture risk index (FRI), calculated on the basis of 1st and 3rd principal strains, and visualizing strain distributions in the proximal femur by generating surrogate FE models via supervised CatBoost model. The training dataset, comprising clinical, anatomical, and mechanical (loading) features, was obtained from QCT image data and QCT-based FEA, which also provided the FRI and strain distributions as targets. The optimized CatBoost model demonstrated 76% accuracy with an AUROC of 0.81 in fracture risk prediction, as well as correlation coefficients of 0.73 for the 1st principal strain and 0.76 for the 3rd principal strain in the surrogate FE models for visualization. Although trained on a limited dataset, this study highlights the efficacy of ML-based surrogate modeling in the QCT-based FEA process for predicting hip fracture risk and visually identifying fracture locations.

Identification of Risk Factors Associated with Early Discontinuation of Physical Therapy Following Shoulder Surgery.

Postoperative physical therapy (POPT) is a critical aspect of patient recovery following orthopedic shoulder surgery. Despite the importance of POPT, some patients do not complete their course of physical therapy, which can affect postoperative recovery and outcomes. This study aimed to identify risk factors that influence patient adherence and attendance of POPT. Patients >90 days postop from elective shoulder surgery were retrospectively recruited for this study from an urban, academic tertiary medical center. Patient charts were reviewed for demographic factors with eligible patients being contacted via telephone call to answer a questionnaire concerning their use of POPT. Inadequate levels of POPT were defined as stopping formal physical therapy earlier than the physical therapist recommended. Final analysis cohorts were compared using Fisher's Exact Test and Wilcoxon Rank Sum Test. A total of 104 patients were included in the final analysis with 84 completing POPT and 20 reporting having stopped therapy early. Patient medical history including elevated BMI (mean significance difference 3.8, [0.018 to 7.6 95% CI]) p=0.026)), being a smoker (p=0.027), having diabetes (p<0.001), and having Medicaid as their primary insurance (p=0.008) were significantly associated with stopping physical therapy earlier than recommended by the physical therapist (Table I). Demographic factors such as requiring additional assistance to get to medical appointments (p=0.012) and living farther away from a physical therapy center (p=0.007) were also substantially associated with a greater risk of stopping physical therapy earlier than recommended by the physical therapist (Table II). Elevated BMI, tobacco use, diabetes, Medicaid enrollment status, requiring additional assistance to get to PT appointments, and living farther away from their physical therapy center are risk factors associated with patients stopping PT after elective shoulder surgery earlier than recommended by a physical therapist. Knowledge of these patient-specific risk factors should be considered by orthopedic surgeons and physical therapists when tailoring a POPT protocol.

A Comparison of Different Strategies for Optimizing the Selection of Empiric Antibiotic Therapy for Pneumonia Caused by Gram-Negative Bacteria in Intensive Care Units: Unit-Specific Combination Antibiograms Versus Patient-Specific Risk Factors.

Guidelines suggest dual antipseudomonal therapy for empiric treatment of pneumonia caused by gram-negative bacteria in intensive care unit (ICU) patients. Additionally, consideration of local susceptibility data and patient-specific risk factors for resistance is recommended for selecting optimal empiric regimens. However, data assessing how to best do this are lacking, and it is unclear whether a local susceptibility data-based or a patient-specific risk factor-based approach will better drive appropriate empiric treatment. This study aims to compare these 2 strategies. This retrospective study was divided into 2 periods. In period I, gram-negative respiratory cultures from ICU patients were used to develop unit-specific combination antibiograms, and individual patient charts were reviewed to assess the impact of risk factors on antimicrobial susceptibility to develop a risk factor-based treatment algorithm. Optimal empiric regimens based on these 2 strategies were then defined. In period II, these regimens were hypothetically applied to patients to compare rates of appropriate empiric therapy and overuse by the 2 methods. Risk factor-based regimens had a higher appropriateness rate compared to regimens derived from antibiograms (89.9% vs 83.7%). Additionally, applying antibiogram-based regimens resulted in a higher prevalence of antibiotic overuse than a patient-specific risk factor-based approach (69.8% vs 40.3%), with excess overuse driven by a higher frequency of unnecessary use of combination therapy. Both strategies provided high rates of appropriateness in empiric antibiotic selection. However, the patient-specific risk factor-based approach demonstrated a higher rate of appropriate therapy and offered advantages in reducing rates of unnecessary combination therapy.

Toward assessment of rupture risk predictors in abdominal aortic aneurysms including intraluminal thrombus based on 3D+t ultrasound images.

Image-based patient-specific rupture risk analysis for abdominal aortic aneurysms (AAAs) has shown considerable promise. However, clinical translation has been hampered by the use of invasive and costly imaging modalities. Despite being a promising alternative, ultrasound (US) makes a full analysis, including intraluminal thrombus (ILT), not trivial. This study explored the feasibility of assessing AAA rupture risk parameters, e.g., peak wall stress (PWS) and peak wall rupture index (PWRI), using US-based models of the AAA wall, finally including ILT. Three-dimensional US data were segmented from a group of AAA patients whose CT data were available within 30 days. The segmented vessel wall and ILT boundaries were converted into a mesh including and excluding ILT to evaluate the effect of adding ILT on the model output. US-based rupture risk parameters (PWS and PWRI) were compared to CT-based results. The US-based PWS and PWRI, including ILT, showed good agreement with CT-based results, and the model excluding ILT showed no significant bias in wall stress or rupture index. When including ILT, a lower US-based wall stress and rupture index of 7.2% and 3.8% were found, respectively. The intraclass correlation coefficient (ICC) of PWS was 0.60. The highest ICC was found for the PWRI (ICC = 0.86), indicating good absolute agreement. This study showed that PWRI can be estimated with US when including the ILT, yielding comparable results to CT, and good absolute agreement. Future work should focus on improving the contrast of ILT in US, since this will be essential to performing large-scale studies in AAA cohorts.

A standards-based software tool to support prognostic modelling: application to exemplar heart failure risk scores

Abstract Introduction Heart failure prognosis is an active research area, with the development of several new risk scores each year. While risk score derivation and evaluation have become common practices, these processes currently lack standardisation. Purpose We aimed to: i) develop a standards-based software tool to support prognostic modelling; and ii) apply the tool to derive, evaluate and compare exemplar heart failure risk models to address a hypothetical question of whether using only non-invasive measurements provides similar prognosis value compared to adding invasive measurements. Methods A software tool was developed based on a prognostic modelling workflow defined using the Common Workflow Language and containing 7 steps: univariate Cox proportional hazard (PH) regression, multivariable Cox PH regression with assessment of proportional hazards and linearity with outcome (Schoenfeld and Martingale residuals), derivation of points per risk factor, risk calculation and stratification, discrimination and calibration assessment. Using data from the PEOPLE heart failure cohort [1], two integer-based risk scores were developed to predict 2-year all-cause mortality. The first model utilised 12 variables available at recruitment informed by prior analysis of the PEOPLE cohort to provide a benchmark for comparison, while the second model is based on 9 non-invasive parameters. The analysis was conducted on cases with complete data (n=781), with no internal or external validation. Results By inputting the dataset and indicating the variables of interest for each model, our software tool automatically generated the patient-specific risk scores, risk strata, c-index, and discrimination and calibration plots. A 12-parameter model based on age, sex, ejection fraction, hypertension, diabetes, atrial fibrillation, ischaemic history, NYHA class, heart rate, systolic blood pressure, creatinine and NT-proBNP yielded reasonable discrimination (c-index=0.77). Graphical assessment of risk strata and calibration (Fig. 1.a and c) revealed good performance overall, apart from the highest risk decile where survival was overestimated. The reduced 9-parameter model excluded creatinine, NT-proBNP, and automatically dropped diabetes status for non-linearity. It showed a substantial drop in discrimination (c-index=0.67), confirmed in graphical assessment of risk strata (Fig. 1.b). Calibration was more variable per decile than seen with the 12-parameter model but was more robust to survival overestimation (Fig. 1.d). Conclusion Our software tool ensures the uniformity and comparability of the two risk models derived by a common statistical process. In this example, we showed how the addition of NT-proBNP, creatinine, and diabetes status improves discrimination for predicting heart failure survival. This tool can be leveraged to test new hypotheses in a standardised and reproducible manner to enhance research in heart failure prognosis.

Deep learning-based segmentation of abdominal aortic aneurysms and intraluminal thrombus in 3D ultrasound images.

Ultrasound (US)-based patient-specific rupture risk analysis of abdominal aortic aneurysms (AAAs) has shown promising results. Input for these models is the patient-specific geometry of the AAA. However, segmentation of the intraluminal thrombus (ILT) remains challenging in US images due to the low ILT-blood contrast. This study aims to improve AAA and ILT segmentation in time-resolved three-dimensional (3D + t) US images using a deep learning approach. In this study a "no new net" (nnU-Net) model was trained on 3D + t US data using either US-based or (co-registered) computed tomography (CT)-based annotations. The optimal training strategy for this low-contrast data was determined for a limited dataset. The merit of augmentation was investigated, as well as the inclusion of low-contrast areas. Segmentation results were validated with CT-based geometries as the ground truth. The model trained on CT-based masks showed the best performance in terms of DICE index, Hausdorff distance, and diameter differences, covering a larger part of the AAA. With a higher accuracy and less manual input the model outperforms conventional methods, with a mean Hausdorff distance of 4.4mm for the vessel and 7.8mm for the lumen. However, visibility of the lumen-ILT interface remains the limiting factor, necessitating improvements in image acquisition to ensure broader patient inclusion and enable rupture risk assessment of AAAs in the future.

Archetypal Analysis of Kidney Allograft Biopsies Using Next-generation Sequencing Technology.

In kidney transplantation, molecular diagnostics may be a valuable approach to improve the precision of the diagnosis. Using next-generation sequencing (NGS), we aimed to identify clinically relevant archetypes. We conducted an Illumina bulk RNA sequencing on 770 kidney biopsies (540 kidney recipients) collected between 2006 and 2021 from 11 European centers. Differentially expressed genes were determined for 11 Banff lesions. An ElasticNet model was used for feature selection, and 4 machine learning classifiers were trained to predict the probability of presence of the lesions. NGS-based classifiers were used in an unsupervised archetypal analysis to different archetypes. The association of the archetypes with allograft survival was assessed using the iBox risk prediction score. The ElasticNet feature selection reduced the number of the genes from a range of 859-10 830 to a range of 52-867 genes. NGS-based classifiers demonstrated robust performances (precision-recall area under the curves 0.708-0.980) in predicting the Banff lesions. Archetypal analysis revealed 8 distinct phenotypes, each characterized by distinct clinical, immunological, and histological features. Although the archetypes confirmed the well-defined Banff rejection phenotypes for T cell-mediated rejection and antibody-mediated rejection, equivocal histologic antibody-mediated rejection, and borderline diagnoses were reclassified into different archetypes based on their molecular signatures. The 8 NGS-based archetypes displayed distinct allograft survival profiles with incremental graft loss rates between archetypes, ranging from 90% to 56% rates 7 y after evaluation (P < 0.0001). Using molecular phenotyping, 8 archetypes were identified. These NGS-based archetypes might improve disease characterization, reclassify ambiguous Banff diagnoses, and enable patient-specific risk stratification.

OrthoMortPred: Predicting one-year mortality following orthopedic hospitalization

ObjectivePredicting mortality risk following orthopedic surgery is crucial for informed decision-making and patient care. This study aims to develop and validate a machine learning model for predicting one-year mortality risk after orthopedic hospitalization and to create a personalized risk prediction tool for clinical use. MethodsWe analyzed data from 3,132 patients who underwent orthopedic procedures at the Central Lisbon University Hospital Center from 2021 to 2023. Using the LightGBM algorithm, we developed a predictive model incorporating various clinical and administrative variables. We employed SHAP (SHapley Additive exPlanations) values for model interpretation and created a personalized risk prediction tool for individual patient assessment. ResultsOur model achieved an accuracy of 93% and an area under the ROC curve of 0.93 for predicting one-year mortality. Notably, ’EMERGENCY ADMISSION DATE TIME’ emerged as the most influential predictor, followed by age and pre-operative days. The model demonstrated robust performance across different patient subgroups and outperformed traditional statistical methods. The personalized risk prediction tool provides clinicians with real-time, patient-specific risk assessments and insights into contributing factors. ConclusionOur study presents a highly accurate model for predicting one-year mortality following orthopedic hospitalization. The significance of ’EMERGENCY ADMISSION DATE TIME’ as the primary predictor highlights the importance of admission timing in patient outcomes. The accompanying personalized risk prediction tool offers a practical means of implementing this model in clinical settings, potentially improving risk stratification and patient care in orthopedic practice.

Determination of Quality Indicators for Microvascular Grafts in Cranio-Maxillofacial Surgery-A Retrospective Analysis of 251 Free Flaps.

The use of microvascular grafts is the gold standard in oral and maxillofacial surgery for the reconstruction of soft tissue and bony and combined defects. Graft loss is one of the most serious complications in the field of reconstructive surgery. A comprehensive analysis of factors influencing this is, therefore, essential. This hypothesis-generating study analyzed 251 patient cases of oral and maxillofacial surgery at the University Hospital Düsseldorf from 2016 to 2020 regarding patient- and therapy-specific parameters for their impact on graft survival. Statistically significant influencing factors were found among the 80 parameters examined: treatment with antiplatelet medication and a BMI ≥ 24.5 at the time of surgery had a positive influence on graft survival, while existing diabetes mellitus, atrial fibrillation, tracheostomy, and a longer operation time had a statistically relevant negative influence. This work demonstrates the relevance of patient-specific risk stratification and the need for further research to develop a valid risk profile. Identifying high-risk patients with medium-sized defects, where alternatives to microvascular reconstruction are available, appears to be crucial for the clinical outcome.

Prediction of brain metastasis development with DNA methylation signatures.

Brain metastases (BMs) are the most common and among the deadliest brain tumors. Currently, there are no reliable predictors of BM development from primary cancer, which limits early intervention. Lung adenocarcinoma (LUAD) is the most common BM source and here we obtained 402 tumor and plasma samples from a large cohort of patients with LUAD with or without BM (n = 346). LUAD DNA methylation signatures were evaluated to build and validate an accurate model predicting BM development from LUAD, which was integrated with clinical factors to provide comprehensive patient-specific BM risk probabilities in a nomogram. Additionally, immune and cell interaction gene sets were differentially methylated at promoters in BM versus paired primary LUAD and had aligning dysregulation in the proteome. Immune cells were differentially abundant in BM versus LUAD. Finally, liquid biomarkers identified from methylated cell-free DNA sequenced in plasma were used to generate and validate accurate classifiers for early BM detection. Overall, LUAD methylomes can be leveraged to predict and noninvasively identify BM, moving toward improved patient outcomes with personalized treatment.

Therapy for Dupuytren’s Disease: Collagenase Therapy—A Long-Term Follow-Up Study

Background: Dupuytren’s disease (DD) is a systemic connective tissue disorder of the palm. It particularly affects men of Northern European or Caucasian origin over the age of 55. In addition to the classical surgical therapy via limited fasciectomy, Dupuytren’s contracture can also be treated minimally invasively. A relatively new treatment method is the use of collagenase injections (Xiapex) to reduce the contracture of the fingers. The data regarding the long-term success of this therapy are currently limited. Methods: In this monocentric retrospective study, we examined 35 patients who were treated with collagenase (Xiapex) for Dupuytren’s contracture in the long fingers. Following the manufacturer’s recommendations, the injection was administered intralesionally, and the cord was ruptured through the passive extension of the finger under local anesthesia with Mepivacain the following day. The clinical follow-up examination was conducted after an average of 5.7 years. The stages of Dupuytren’s disease were documented using the Tubiana classification. Additionally, parameters of finger extension ability, differentiated by metacarpophalangeal (MCP), and proximal interphalangeal (PIP) joints, as well as patient-specific risk parameters, were evaluated Results: The long-term results of collagenase therapy after an average of 5.7 years showed a significant improvement in the contracture of the affected fingers. In the MCP joints, the flexion contracture decreased from 42° to 17° (p ≤ 0.001), and in the PIP joints, it decreased from 56° to 33° (p ≤ 0.001). The primary recurrence rate was 11% for the MCP joints and 19% for the PIP joints, respectively. The analysis of risk factors showed a significant risk for worse long-term outcomes in patients with diabetes and those with nicotine abuse. Conclusions: Collagenase therapy for Dupuytren’s disease achieved significant long-term improvements in contracture in both MCP and PIP joints. In accordance with general risk factors for DD, patients with diabetes and those with nicotine abuse are at risk of worse long-term outcomes. Overall, it is a time-saving, low-risk, and straightforward technique for treating the disabling contracture component of this disease.