University Surgical Associates

<title>Abstract</title> Background: Amyotrophic Lateral Sclerosis (ALS) presents significant challenges for biomarker identification and brain-computer interface (BCI) development due to progressive motor neuron degeneration and high signal-to-noise ratios in neurophysiological recordings. Traditional EEG analysis methods struggle with patient-specific variability and environmental artifacts, limiting their clinical utility. Methods: We present an adaptive hybrid framework integrating dendritic-inspired noise filtering, quantum-inspired symbolic search with Hamming distance optimization, and meta-adaptive machine learning (MAML) for personalized adaptation. The system processes synchronized EEG and eye-tracking data from the EEGET-ALS dataset (45 ALS patients, 42 healthy controls), implementing multiple threshold optimization strategies including Bayesian optimization, evolutionary algorithms, and ensemble methods. The framework maintains a historical performance database enabling continuous learning and transfer learning capabilities across patient cohorts. Results: Our framework achieved an F1-score of 0.967 (95% CI: 0.958-0.976), precision of 0.98, and recall of 0.95 in distinguishing ALS patients from healthy controls (Figure 1o). Key neurophysiological findings include ~25% beta power reduction (13-30 Hz) in central electrodes (Figure 1a-c), 40% motor connectivity loss particularly in C3-C4 coupling (Figure 1e-g), and preserved gamma oscillatory patterns (Figure 1d). The meta-learning module improved performance from baseline F1=0.667 to F1=0.751 within 1000 search iterations (Figure 1h), with transfer learning accelerating convergence by 5-10x (Figure 1l). Conclusions: The quantum-inspired adaptive framework establishes a new paradigm for ALS biomarker discovery and BCI development. By combining biologically-inspired filtering with continuous meta-learning, the system achieves clinical-grade accuracy while adapting to patient-specific variations. This approach enables robust communication interfaces for ALS patients and provides a generalizable framework for neurological biomarker discovery.

In a time of instantaneous decision making, streaming data has become an essential tool for customer analytics.The reality is that organizations can leverage real-time insights if they can make sense of the vast, fast data coming from digital and social platforms, IoT devices, and transactional systems.However, making the best use of this data requires machine learning (ML) models that are both accurate and interpretable.As soon as explainable artificial intelligence (XAI) started to gain traction, demand for models able to provide interpretability and justifiability increased even further, particularly in consumer-facing domains characterized by a high degree of reliance on trust, fairness, and regulation.We present a framework for explainable customer analytics in streaming systems, based on the use of interpretable machine learning models.In contrast to traditional batch-learning techniques, our method focuses on low-latency prediction, model adaptability, and human-understandable insights for streaming data.Chapter 2 explains the following customer analytics tasks: churn prediction, segmentation, personalization, lifetime value estimation, and the challenges associated with real-time data processing.We consider recent interpretable ML approaches, such as decision trees, rule-based classifiers, monotonic gradient boosting, and model-agnostic post-hoc explanations (e.g., LIME and SHAP).These approaches are analysed in terms of their suitability for streaming architectures and computational complexity.The paper concludes with a data engineering pipeline implemented on top of Apache Kafka and Apache Flink, which collects and preprocesses an online training dataset, then serves real-time data to the lightweight, interpretable models.We also use concept drift detection techniques to ensure the model's relevance over time.To verify the proposed method, we conducted experiments using publicly available customer data and a simulation of streaming with augmented customer data.The performance of models is evaluated not only on prediction accuracy, but also on interpretability measures, including model fidelity, coverage, and stability.The findings indicate that tree-based models supplemented with SHAP explanations achieve an acceptable trade-off between (real-time) performance and interpretability.We also discuss case studies where the model recommendations are transformed into business actions, such as delivering retention incentives or updating product recommendations, and demonstrate how interpretability strengthens trust among stakeholders and enhances operational efficiency.Although we leave these studies out of scope, in these works, ethical and regulatory considerations of nontransparent ML (including in the case of customer analytics in GDPR, CCPA, and the future AI Act) are considered in detail.Explainability enables organizations to provide customers with meaningful explanations for automated decisions, thereby mitigating legal risk and enhancing the acceptance of AI-driven insights.A roadmap is developed to describe how to integrate explainable machine learning with enterprise-level customer analytics platforms.We provide guidance on model selection, explanatory tools, streaming processing infrastructure, and organizational governance.Our results suggest that a shift from black-box optimization to a more transparent and responsible AI in customer analytics is warranted when decisions have a significant impact on consumer experience and loyalty.

Introduction: As the field of cosmetic and plastic surgery progresses patient safety and comfort remain paramount. Over the past 30 years, multiple studies have demonstrated the effectiveness of progressive tension sutures (PTS) in drainless abdominoplasties and case series have been published evaluating the outcomes of PTS/drainless abdominoplasties in high-volume liposuction cases (liposuction with over 3 L of aspirate). To date, there have been no reports in the literature regarding PTS usage in Fleur de Lis (FDL)-type abdominoplasties. Materials/Methods: We describe our experience using PTS in a patient with a drainless FDL abdominoplasty and several patients in which PTS was used to limit the number of drains needed in an FDL abdominoplasty. Results: In our follow-up with the patient, she had no adverse events or seroma formation. Discussion/conclusion: The cases documented here suggest that FDL abdominoplasty can be performed without the need for drains or with minimal drains.

Neuroimmune interactions are essential for the development of neuropathic pain, yet the contributions of distinct immune cell populations have not been fully unraveled. Here, we demonstrate the critical role of B cells in promoting mechanical hypersensitivity (allodynia) after peripheral nerve injury in male and female mice. Depletion of B cells with a single injection of anti-CD20 monoclonal antibody at the time of injury prevented the development of allodynia. B cell-deficient (muMT) mice were similarly spared from allodynia. Nerve injury was associated with increased immunoglobulin G (IgG) accumulation in ipsilateral lumbar dorsal root ganglia (DRGs) and dorsal spinal cords. IgG was colocalized with sensory neurons and macrophages in DRGs and microglia in spinal cords. IgG also accumulated in DRG samples from human donors with chronic pain, colocalizing with a marker for macrophages and satellite glia. RNA sequencing revealed a B cell population in naive mouse and human DRGs. A B cell transcriptional signature was enriched in DRGs from human donors with neuropathic pain. Passive transfer of IgG from injured mice induced allodynia in injured muMT recipient mice. The pronociceptive effects of IgG are likely mediated through immune complexes interacting with Fc gamma receptors (FcγRs) expressed by sensory neurons, microglia, and macrophages, given that both mechanical allodynia and hyperexcitability of dissociated DRG neurons were abolished in nerve-injured FcγR-deficient mice. Consistently, the pronociceptive effects of IgG passive transfer were lost in FcγR-deficient mice. These data reveal that a B cell-IgG-FcγR axis is required for the development of neuropathic pain in mice.

Background Stroke is the leading cause of adult disability worldwide, with approximately 30% of strokes remaining cryptogenic. One potential etiology is a patent foramen ovale (PFO), which may contribute to stroke through paradoxical thrombo-embolism or in situ thrombo-embolus formation. Recent advancements in robot-assisted transcranial Doppler (raTCD) have shown increased sensitivity in detecting right-to-left shunt (RLS) compared to transthoracic echocardiography (TTE), particularly in detecting large shunts, which are associated with a higher risk of stroke. Methods We conducted a retrospective quality improvement project at our comprehensive stroke center to compare the performance of TTE and raTCD in identifying RLS in ischemic stroke patients. The study involved 148 patients admitted between February 2021 and February 2023. All patients underwent TTE and raTCD with agitated saline bubble contrast, with additional transesophageal echocardiography (TEE) at the treatment team's discretion. The primary metrics analyzed included differences in overall RLS detection and large RLS detection rates for raTCD, TTE and TEE. Results raTCD detected RLS in 60.1% of patients compared to 37.2% with TTE (p = 0.00012), with a 42.6% detection rate for large shunts on raTCD versus 23.0% on TTE (p = 0.00053). The sensitivity and specificity of raTCD were 92% and 87.5%, respectively, compared to 78.57% and 71.43% for TTE, using TEE as the gold standard. Nine patients underwent PFO closure, all correctly identified with large shunts by raTCD, while TTE missed or underestimated the PFO size in 44% of the cases. Conclusion raTCD significantly outperforms TTE in detecting RLS and large shunts, suggesting its integration into standard PFO workup protocols may enhance secondary stroke prevention. These findings support the adoption of raTCD as a complementary diagnostic tool alongside TTE and TEE for more accurate PFO detection and risk stratification.

Background: Hypertonic saline (HTS) is essential pharmacologic treatment for traumatic brain injury (TBI). Previous studies associate HTS with acute kidney injury (AKI), however evidence in TBIs is limited. Objective: This study examines factors associated with AKI in patients requiring HTS for TBI. Methods: This retrospective study was performed at a Level-1 Trauma, Academic Medical Center. Inclusion criteria were TBI, age ≥12 years, ICU length of stay ≥72 hours, and administration of ≥24 hours of continuous HTS or 500 mL of HTS boluses. The primary outcome was identifying factors associated with AKI. Secondary outcomes included correlation between chloride load and level with development of AKI. Chloride load was calculated from HTS and non-HTS sources. Results: Of 129 patients included, 18 (14%) developed AKI. Maximum sodium level was higher in the AKI group (P < 0.0001). Hyperchloremia (Cl ≥ 115 mEq/L) was more common in the AKI group (100% vs 81%, P = 0.0428). Maximum and change in serum chloride were higher in the AKI group (median 128 vs 123 mEq/L, P = 0.0026 and +24 mEq/L vs +17 mEq/L, P = 0.0084, respectively). Logistic regression analysis indicated an OR 1.095 times higher [95% CI (1.022, 1.172)] for developing AKI for every one mEq/L increase in maximum chloride level and an OR 1.032 [95% CI (1.006, 1.058)] for developing AKI for every 1-year increase in age. There was no difference in total chloride load between groups (P = 0.2143). Non-HTS sources provided more than 40% of total chloride load in both groups. Conclusion: Chloride level, and age may be associated with AKI in TBI patients treated with HTS.

BackgroundFew studies have evaluated the use of laparoscopic staplers in robotic procedures (bedside stapling, BS). This study aims to evaluate the effectiveness of BS compared with robotic staplers (RS) in bariatric robotic procedures.MethodsPatients who underwent robotic sleeve gastrectomy or gastric bypass elective procedures between 1/1/2021 and 12/31/2021 were extracted from PINC AI™ Healthcare Data. The following clinical outcomes were compared: blood transfusion, bleeding, anastomotic leak, intensive care unit (ICU) visit, and 30-day readmission, operating room (OR) time, inpatient costs, and length of stay. We evaluated baseline balance in BS and RS and bivariate association between covariates and outcomes using Chi-square or Fisher exact test and t-test or ANOVA. Multivariable general linear mixed models (GLMMs) with respective gamma or binomial distribution and log-link function were used to obtain adjusted outcomes variations between BS and RS.ResultsTotal of 7268 discharges were included with 1603 (22.1%) BS and 5665 (77.9%) RS cases. RS cases consisted of a higher number of patients who were Hispanic (17.0% vs. 9.4%), had Medicaid (26.9% vs. 19.4%) and underwent sleeve gastrectomy (68.4% vs. 53.5%). Higher proportions of RS cases were done by providers in Northeast region (35.5% vs. 24.3%), smaller size (< 500 beds; 71.1% vs. 52.3%), and teaching hospitals (59.4% vs. 39%). The adjusted outcomes variations demonstrated that patients that had RS were significantly more likely to have blood transfusions, ICU stays, increased ORT (19 min) and costs ($1273). Sensitivity analysis showed similar results, except no significant differences in blood transfusion rates in both groups.ConclusionsBedside staplers significantly reduce healthcare resource utilization with equivalent effectiveness and fewer ICU stays compared to robotic staplers.Graphical abstract