Clinical Applications of Machine Learning.

Objective: This review introduces interpretable predictive machine learning approaches, natural language processing, image recognition, and reinforcement learning methodologies to familiarize end users. Background: As machine learning, artificial intelligence, and generative artificial intelligence become increasingly utilized in clinical medicine, it is imperative that end users understand the underlying methodologies. Methods: This review describes publicly available datasets that can be used with interpretable predictive approaches, natural language processing, image recognition, and reinforcement learning models, outlines result interpretation, and provides references for in-depth information about each analytical framework. Results: This review introduces interpretable predictive machine learning models, natural language processing, image recognition, and reinforcement learning methodologies. Conclusions: Interpretable predictive machine learning models, natural language processing, image recognition, and reinforcement learning are core machine learning methodologies that underlie many of the artificial intelligence methodologies that will drive the future of clinical medicine and surgery. End users must be well versed in the strengths and weaknesses of these tools as they are applied to patient care now and in the future.

Survival Tree Provides Individualized Estimates of Survival After Lung Transplant.

INTRODUCTION: Identifying contributors to lung transplant survival is vital in mitigating mortality. To enhance individualized mortality estimation and determine variable interaction, we employed a survival tree algorithm utilizing recipient and donor data. METHODS: United Network Organ Sharing data (2000-2021) were queried for single and double lung transplants in adult patients. Graft survival time <7 d was excluded. Sixty preoperative and immediate postoperative factors were evaluated with stepwise logistic regression on mortality; final model variables were included in survival tree modeling. Data were split into training and testing sets and additionally validated with 10-fold cross validation. Survival tree pruning and model selection was based on Akaike information criteria and log-likelihood values. Estimated survival probabilities and log-rank pairwise comparisons between subgroups were calculated. RESULTS: A total of 27,296 lung transplant patients (8175 single; 19,121 double lung) were included. Stepwise logistic regression yielded 47 significant variables associated with mortality. Survival tree modeling returned six significant factors: recipient age, length of stay from transplant to discharge, recipient ventilator duration post-transplant, double lung transplant, recipient reintubation post-transplant, and donor cytomegalovirus status. Eight subgroups consisting of combinations of these factors were identified with distinct Kaplan-Meier survival curves. CONCLUSIONS: Survival trees provide the ability to understand the effects and interactions of covariates on survival after lung transplantation. Individualized survival probability with this technique found that preoperative and postoperative factors influence survival after lung transplantation. Thus, preoperative patient counseling should acknowledge a degree of uncertainty given the influence of postoperative factors.

Statewide Hospital Admissions for Adult Survivors of Infant Surgical Diseases Over a 10-Year Period.

INTRODUCTION: The number of patients with congenital disease living to adulthood continues to grow. Often undergoing surgical correction in infancy, they continue to require lifelong care. Their numbers are largely unknown. We sought to evaluate hospital admissions of adult patients with esophageal atresia with tracheoesophageal fistula (EA/TEF), congenital diaphragmatic hernia (CDH), and Hirschsprung disease (HD). METHODS: The Florida Agency for Healthcare Administration inpatient database was merged with the Distressed Communities Index and Centers for Medicare and Medicaid Services Hospital and Physician Compare datasets. The dataset was queried for adult patients (≥18 y, born after 1970) with EA/TEF, CDH, and HD in their problem list from 2010 to 2020. Patient demographics, hospitalization characteristics, and discharge information were obtained. RESULTS: In total, 1140 admissions were identified (266 EA/TEF, 135 CDH, 739 HD). Patients were mostly female (53%), had a mean age of 31.6 y, and often admitted to an adult internist in a general hospital under emergency. Principal diagnoses and procedures (when performed) varied with diagnosis and age at admission. EA patients were admitted with dysphagia and foregut symptoms and often underwent upper endoscopy with dilation. CDH patients were often admitted for diaphragmatic hernias and underwent adult diaphragm repair. Hirschsprung patients were often admitted for intestinal obstructive issues and frequently underwent colonoscopy but trended toward operative intervention with increasing age. CONCLUSIONS: Adults with congenital disease continue to require hospital admission and invasive procedures. As age increases, diagnoses and performed procedures for each diagnoses evolve. These data could guide the formulation of multispecialty disease-specific follow-up programs for these patients.

Meta-Analysis in Surgical Research: Methodology and Statistical Application.

In evidence-based medicine, systematic review continues to carry the highest weight in terms of quality and reliability, synthesizing robust information from previously published cohort studies to provide a comprehensive overview of a topic. Meta-analysis provides further depth by allowing for comparative analysis between the studied intervention and the control group, providing the most up-to-date evidence on their characteristics and efficacy. We discuss the principles and methodology of meta-analysis, and its applicability to the field of surgical research. The clinical question is defined using PICO framework (Problem, Intervention, Comparison, Outcome). Then a systematic article search is performed across multiple medical databases using relevant search terms, which are then filtered out based on appropriate screening tools. Pertinent data from the selected articles are collected and undergo critical appraisal by at least two independent reviewers. Additional statistical tests may be performed to identify the presence of any significant bias. The data are then synthesized to perform comparative analysis between the intervention and comparison groups. In this article, we discuss specifically the usage of R software (R Foundation for Statistical Computing, Vienna, Austria) for data analysis and visualization. Meta-analysis results of the pooled data are presented using forest plots. Concerns for potential bias may be addressed through the creation of funnel plots. Meta-analysis is a powerful tool to provide highly reliable medical evidence. It may be readily performed by independent researchers with minimal need for funding or institutional approval. The ability to conduct such studies is an asset to budding medical scholars.

Interfacial crystallized oleogel emulsion with improved freeze-thaw stability and tribological properties: Influence of cooling rate.

In this study, the influence of cooling rate on the freeze-thaw stability, rheological and tribological properties of interfacial crystalized oleogel emulsion was investigated. Results showed that slower cooling rate could promote formation of larger crystals and stronger network in oleogels. Additionally, oleogel emulsions showed higher freeze-thaw stability than those stabilized solely by emulsifiers. The slower cooling rate resulted in larger crystals adsorbed at the droplet surface. This led to greater steric hindrance that prevented the migration of oil droplets with higher resistance to disruption by ice crystals. The rheological and tribological measurements suggested that with appropriate amount of crystals, the tribological properties were better maintained for emulsions prepared at slow cooling rate after freeze-thaw treatment. This strategy greatly enriched oleogel emulsion formulations and provided important clues for potential applications in food products involved with freeze-thaw treatment.

Faster discharge with lactated ringers than normal saline in first 72 h of acute pancreatitis: A multicenter randomized trial.

OBJECTIVES: Data driven strategies for acute pancreatitis (AP) in pediatrics are limited; adult data suggests lactated ringers (LR) compared to normal saline (NS) resulted in favorable outcomes, but has not been studied in pediatrics. Our objective was to evaluate the efficacy of LR during the first 48 h of an AP episode compared with NS. STUDY DESIGN: A multisite randomized controlled clinical trial, from 2015 to 2020 (Clinical Trials.gov NCT03242473). Patients were randomized to exclusively LR or NS for the first 48 h. Primary outcomes were serial C-reactive protein (CRP) values. Secondary outcomes included other lab values, time to feeds, length of stay (LOS), systemic inflammatory response syndrome (SIRS) development, and progression to severe AP (SAP). RESULTS: We studied 76 patients (38 LR, 38 NS). CRP at 24 and 48 h were not significantly different between LR or NS group. Additionally, there were no differences in trends of BUN, amylase, lipase, SIRS status, or SAP development between the LR and NS group at 24 and 48 h. A higher proportion of LR patients (32%, 12/38) were discharged before 48 h compared to NS (13%, 5/38). The LR group had a significantly higher rate of discharge within the first 72 h compared to the NS group (p = 0.02). CONCLUSION: The use of LR was associated with a faster rate of discharge during the intervention period and in the first 72 h, but no other differences compared to NS. This reduction in length of hospitalization has significant implications for patients and healthcare costs.

Single-Position Prone Lateral Lumbar Interbody Fusion Technique Guide: Surgical Tips and Tricks.

Lateral lumbar interbody fusion (LLIF) is a popular technique as it allows for the placement of a large interbody implant through a retroperitoneal, transpsoas working corridor. Historically, the interbody is placed with the patient in lateral decubitus and then repositioned to prone for the posterior instrumentation. While this has been an effective and successful technique, removing the interoperative flip would improve the efficiency of these cases. This has led to modified LLIF approaches including single-position prone LLIF (pLLIF). This modification has shown to be an efficient and powerful technique; however, learning to navigate the LLIF approach in the prone position has its own challenges. The purpose of this article is to provide a detailed description of our pLLIF technique while simultaneously introducing surgical tips to overcome the challenges of the approach and optimize the implantation of the interbody device.

Comparison of complication rates between anterior versus posterior approaches for treating unstable Hangman's fracture. A systematic review and meta-analysis.

Study design: Systematic Review and Meta-analysis. Objective: To compare the complication rates associated with anterior and posterior approaches for the surgical treatment of unstable hangman's fractures. Methods: A systematic review and meta-analysis were performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines in PubMed, Web of Science, and Scopus databases to identify comparative studies reporting complications of anterior versus posterior approaches for the treatment of unstable hangman's fractures. Results: The search yielded 1163 papers from which 5 studies were fully included. One hundred fifteen (115) patients were operated on using an anterior approach versus 65 through a posterior approach. The average complication rates for the anterior and posterior approaches were 26.1 % and 13.8 %, respectively. No complications following the anterior approach required pharmacological or surgical intervention (Clavien-Dindo, Grade 1), while 88.9 % of complications following the posterior approach did (Clavien-Dindo, Grade 2). Conclusion: No significant differences in the complication rates were found when comparing anterior versus posterior surgery for treating a C2 traumatic spondylolisthesis. However, most of the complications presented in the posterior surgery group were more severe.

Is nonoperative management of appendicitis inferior in the COVID-19 era?

BACKGROUND: Previous studies on nonoperative management (NOM) of acute appendicitis (AA) indicated comparable outcomes to surgery, but the effect of COVID-19 infection on appendicitis outcomes remains unknown. Thus, we evaluate appendicitis outcomes during the COVID-19 pandemic to determine the effect of COVID-19 infection status and treatment modality. We hypothesized that active COVID-19 patients would have worse outcomes than COVID-negative patients, but that outcomes would not differ between recovered COVID-19 and COVID-negative patients. Moreover, we hypothesized that outcomes would not differ between nonoperative and operative management groups, regardless of COVID-19 status. METHODS: We queried the National COVID Cohort Collaborative from 2020 to 2023 to identify adults with AA who underwent operative or NOM. COVID-19 status was denoted as follows: COVID-negative, COVID-active, or COVID-recovered. Intention to treat was used for NOM. Propensity score-balanced analysis was performed to compare outcomes within COVID groups, as well as within treatment modalities. RESULTS: A total of 37,868 patients were included: 34,866 COVID-negative, 2,540 COVID-active, and 460 COVID-recovered. COVID-active and recovered less often underwent operative management. Unadjusted, there was no difference in mortality between COVID groups for operative management. There was no difference in rate of failure of NOM between COVID groups. Adjusted analysis indicated, compared with operative, NOM carried higher odds of mortality and readmission for COVID-negative and COVID-active patients. CONCLUSION: This study demonstrates higher odds of mortality among NOM of appendicitis and near equivalent outcomes for operative management regardless of COVID-19 status. We conclude that NOM of appendicitis is associated with worse outcomes for COVID-active and COVID-negative patients. In addition, we conclude that a positive COVID test or recent COVID-19 illness alone should not preclude a patient from appendectomy for AA. Surgeon clinical judgment of a patient's physiology and surgical risk should, of course, inform the decision to proceed to the operating room. LEVEL OF EVIDENCE: Therapeutic/Care Management; Level III.

Artificial Intelligence in Surgical Research: Accomplishments and Future Directions.

MINI-ABSTRACT: The study introduces various methods of performing conventional ML and their implementation in surgical areas, and the need to move beyond these traditional approaches given the advent of big data. OBJECTIVE: Investigate current understanding and future directions of machine learning applications, such as risk stratification, clinical data analytics, and decision support, in surgical practice. SUMMARY BACKGROUND DATA: The advent of the electronic health record, near unlimited computing, and open-source computational packages have created an environment for applying artificial intelligence, machine learning, and predictive analytic techniques to healthcare. The "hype" phase has passed, and algorithmic approaches are being developed for surgery patients through all stages of care, involving preoperative, intraoperative, and postoperative components. Surgeons must understand and critically evaluate the strengths and weaknesses of these methodologies. METHODS: The current body of AI literature was reviewed, emphasizing on contemporary approaches important in the surgical realm. RESULTS AND CONCLUSIONS: The unrealized impacts of AI on clinical surgery and its subspecialties are immense. As this technology continues to pervade surgical literature and clinical applications, knowledge of its inner workings and shortcomings is paramount in determining its appropriate implementation.

Wheat milling across history altered sugar bioaccessibility assessed using TIM-1 in vitro digestion model.

The differences in wheat flour characteristics caused by ancient (pestle and mortar), old (stone hand mill), and modern (roller and cyclone) milling techniques and their effect on in vitro starch digestibility of wheat porridge using the simulated TIM Gastrointestinal Model (TIM-1) were investigated. Ancient flour (AF) was the coarsest flour (∼70 % is >1000 µm), followed by old wholemeal flour (OWF) and old refined flour (ORF) with similar particle size distribution showing one prominent peak (at ∼1000 µm for OWF and ∼800 µm for ORF). Modern refined flour (MRF) had a monomodal distribution centered at a particle size of ∼100 µm, while modern wholemeal flour (MWF) particle size was distributed between 40 and 600 µm. MRF and MWF porridges had higher cumulative sugar bioaccessibility than OWF and AF porridges, with ORF porridge having an intermediate cumulative sugar bioaccessibility. Characterizing the cumulative sugar bioaccessibility profile with a shifted logistic model allows identifying that the maximum sugar bioaccessibility and rate of sugar release were significantly higher (p < 0.05) for MRF and MWF compared to OWF and AF porridges, while the induction times were shorter, demonstrating the importance of processing on modulating starch digestibility.

KDM5-mediated activation of genes required for mitochondrial biology is necessary for viability in Drosophila.

Histone-modifying proteins play important roles in the precise regulation of the transcriptional programs that coordinate development. KDM5 family proteins interact with chromatin through demethylation of H3K4me3 as well as demethylase-independent mechanisms that remain less understood. To gain fundamental insights into the transcriptional activities of KDM5 proteins, we examined the essential roles of the single Drosophila Kdm5 ortholog during development. KDM5 performs crucial functions in the larval neuroendocrine prothoracic gland, providing a model to study its role in regulating key gene expression programs. Integrating genome binding and transcriptomic data, we identify that KDM5 regulates the expression of genes required for the function and maintenance of mitochondria, and we find that loss of KDM5 causes morphological changes to mitochondria. This is key to the developmental functions of KDM5, as expression of the mitochondrial biogenesis transcription factor Ets97D, homolog of GABPα, is able to suppress the altered mitochondrial morphology as well as the lethality of Kdm5 null animals. Together, these data establish KDM5-mediated cellular functions that are important for normal development and could contribute to KDM5-linked disorders when dysregulated.

Interpretable machine learning accurately reclassifies lobectomy surgical approaches by cost.

BACKGROUND: The volume of robotic lung resection continues to increase despite its higher costs and unproven superiority to video-assisted thoracoscopic surgery. We evaluated whether machine learning can accurately identify factors influencing cost and reclassify high-cost operative approaches into lower-cost alternatives. METHODS: The Florida Agency for Healthcare Administration and Centers for Medicare and Medicaid Services Hospital and Physician Compare datasets were queried for patients undergoing open, video-assisted thoracoscopic surgery and robotic lobectomy. K-means cluster analysis was used to identify robotic clusters based on total cost. Predictive models were built using artificial neural networks, Support Vector Machines, Classification and Regression Trees, and Gradient Boosted Machines algorithms. Models were applied to the high-volume robotic group to determine patients whose cost cluster changed if undergoing a video-assisted thoracoscopic surgery approach. A local interpretable model-agnostic explanation approach personalized cost per patient. RESULTS: Of the 6,618 cases included in the analysis, we identified 4 cost clusters. Application of artificial neural networks to the robotic subgroup identified 1,642 (65%) cases with no re-assignment of cost cluster, 583 (23%) with reduced costs, and 300 (12%) with increased costs if they had undergone video-assisted thoracoscopic surgery approach. The 5 overall highest cost predictors were patient admission from the clinic, diagnosis of metastatic cancer, presence of cancer, urgent hospital admission, and dementia. CONCLUSION: K-means cluster analysis and machine learning identify a patient population that may undergo video-assisted thoracoscopic surgery or robotic lobectomy without a significant difference in total cost. Local interpretable model-agnostic explanation identifies individual patient factors contributing to cost. Application of this modeling may reliably stratify high-cost patients into lower-cost approaches and provide a rationale for reducing expenditure.

Seasonal Patterns in the Frequency of Candidatus Liberibacter Asiaticus in Populations of Diaphorina citri (Hemiptera: Psyllidae) in Florida.

Candidatus Liberibacter asiaticus (CLas) is one of the putative causal agents of huanglongbing, which is a serious disease in citrus production. The pathogen is transmitted by Diaphorina citri Kuwayama (Hemiptera: Psyllidae). As an observational study, six groves in central Florida and one grove at the southern tip of Florida were sampled monthly from January 2008 through February 2012 (50 months). The collected psyllids were sorted by sex and abdominal color. Disease prevalence in adults peaked in November, with a minor peak in February. Gray/brown females had the highest prevalence, and blue/green individuals of either sex had the lowest prevalence. CLas prevalence in blue/green females was highly correlated with the prevalence in other sexes and colors. Thus, the underlying causes for seasonal fluctuations in prevalence operated in a similar fashion for all psyllids. The pattern was caused by larger nymphs displacing smaller ones from the optimal feeding sites and immunological robustness in different sex-color morphotypes. Alternative hypotheses were also considered. Improving our understanding of biological interactions and how to sample them will improve management decisions. We agree with other authors that psyllid management is critical year-round.

Left ventricular assist device exchange: a review of indications, operative procedure, and outcomes.

The use of left ventricular assist devices (LVADs) is intended to treat patients with end-stage heart failure. Owing to technological advances, these devices are becoming more durable. However, LVADs may need to be exchanged when complications arise and heart transplantation is not possible. Indications for LVAD exchange (LVADE) include device thrombosis, device infections, and pump component failure. LVADE has historically been associated with a high risk of morbidity and mortality. In this review, we discuss the indications of LVADE, the decisional and technical aspects during surgery, and outcomes.

Therapeutic activity of lipoxin A4 in TiO2-induced arthritis in mice: NF-κB and Nrf2 in synovial fluid leukocytes and neuronal TRPV1 mechanisms.

Background: Lipoxin A4 (LXA4) has anti-inflammatory and pro-resolutive roles in inflammation. We evaluated the effects and mechanisms of action of LXA4 in titanium dioxide (TiO2) arthritis, a model of prosthesis-induced joint inflammation and pain. Methods: Mice were stimulated with TiO2 (3mg) in the knee joint followed by LXA4 (0.1, 1, or 10ng/animal) or vehicle (ethanol 3.2% in saline) administration. Pain-like behavior, inflammation, and dosages were performed to assess the effects of LXA4 in vivo. Results: LXA4 reduced mechanical and thermal hyperalgesia, histopathological damage, edema, and recruitment of leukocytes without liver, kidney, or stomach toxicity. LXA4 reduced leukocyte migration and modulated cytokine production. These effects were explained by reduced nuclear factor kappa B (NFκB) activation in recruited macrophages. LXA4 improved antioxidant parameters [reduced glutathione (GSH) and 2,2-azino-bis 3-ethylbenzothiazoline-6-sulfonate (ABTS) levels, nuclear factor erythroid 2-related factor 2 (Nrf2) mRNA and Nrf2 protein expression], reducing reactive oxygen species (ROS) fluorescent detection induced by TiO2 in synovial fluid leukocytes. We observed an increase of lipoxin receptor (ALX/FPR2) in transient receptor potential cation channel subfamily V member 1 (TRPV1)+ DRG nociceptive neurons upon TiO2 inflammation. LXA4 reduced TiO2-induced TRPV1 mRNA expression and protein detection, as well TRPV1 co-staining with p-NFκB, indicating reduction of neuronal activation. LXA4 down-modulated neuronal activation and response to capsaicin (a TRPV1 agonist) and AITC [a transient receptor potential ankyrin 1 (TRPA1) agonist] of DRG neurons. Conclusion: LXA4 might target recruited leukocytes and primary afferent nociceptive neurons to exert analgesic and anti-inflammatory activities in a model resembling what is observed in patients with prosthesis inflammation.

Pediatric Intestinal Transplantation Management and Outcomes.

Intestinal failure is the anatomic or functional loss of intestinal function below the minimum required to absorb nutrients to maintain health and growth. Parenteral nutrition is the main supportive therapy for children with intestinal failure, but if serious complications develop, intestinal transplantation may be needed to sustain life. Referral to a multidisciplinary intestinal rehabilitation team and an extensive evaluation are necessary steps before listing for transplantation. Immunosuppression is part of life-long therapy after transplantation, and children continue to have high medical needs. Serious complications include acute cellular rejection, graft-versus-host disease, infection, and post-transplant lymphoproliferative disease. However, intestinal transplantation has led to improved outcomes in recent years and is a viable life-saving option for many children with intestinal failure.

Machine Learning Analysis of Postlaparoscopy Hernias and "I'm Leaving You to Close" Strategy.

INTRODUCTION: Contributing factors to postlaparoscopy hernia are unknown. We hypothesized that postlaparoscopy incisional hernias are increased when the index surgery was performed in teaching hospitals. Laparoscopic cholecystectomy was chosen as the archetype for open umbilical access. MATERIALS AND METHODS: Maryland and Florida SID/SASD databases (2016-2019) wereused to track 1-year hernia incidence in both inpatient and outpatient settings, which was then linked to Hospital Compare, Distressed Communities Index (DCI), and ACGME. Postoperative umbilical/incisional hernia following laparoscopic cholecystectomy was identified using CPT and ICD-10. Propensity matching and eight machine learning modes were utilized including logistic regression, neural network, gradient boosting machine, random forest, gradient boosted trees, classification and regression trees, k nearest neighbors and support vector machines. RESULTS: Postoperative hernia incidence was 0.2% (total = 286; 261 incisional and 25 umbilical) in 117,570 laparoscopic cholecystectomy cases. Days to presentation (mean ± SD) were incisional 141 ± 92 and umbilical 66 ± 74. Logistic regression performed best (AUC 0.75 (95% ci 0.67-0.82) and accuracy 0.68 (95% ci 0.60-0.75) using 10-fold cross validation) in propensity matched groups (1:1; n = 279). Postoperative malnutrition (OR 3.5), hospital DCI of comfortable, mid-tier, at risk or distressed (OR 2.2 to 3.5), LOS >1 d (OR 2.2), postop asthma (OR 2.1), hospital mortality below national average (OR 2.0) and emergency admission (OR 1.7) were associated with increased hernias. A decreased incidence was associated with patient location of small metropolitan areas with <1 million residents (OR 0.5) and Charlson Comorbidity Index-Severe (OR 0.5). Teaching hospitals were not associated with postoperative hernia after laparoscopic cholecystectomy. CONCLUSIONS: Different patient factors as well as underlying hospital factors are associated with postlaparoscopy hernias. Performance of laparoscopic cholecystectomy at teaching hospitals is not associated with increased postoperative hernias.

Vaccines Alone Cannot Slow the Evolution of SARS-CoV-2.

The rapid emergence of immune-evading viral variants of SARS-CoV-2 calls into question the practicality of a vaccine-only public-health strategy for managing the ongoing COVID-19 pandemic. It has been suggested that widespread vaccination is necessary to prevent the emergence of future immune-evading mutants. Here, we examined that proposition using stochastic computational models of viral transmission and mutation. Specifically, we looked at the likelihood of emergence of immune escape variants requiring multiple mutations and the impact of vaccination on this process. Our results suggest that the transmission rate of intermediate SARS-CoV-2 mutants will impact the rate at which novel immune-evading variants appear. While vaccination can lower the rate at which new variants appear, other interventions that reduce transmission can also have the same effect. Crucially, relying solely on widespread and repeated vaccination (vaccinating the entire population multiple times a year) is not sufficient to prevent the emergence of novel immune-evading strains, if transmission rates remain high within the population. Thus, vaccines alone are incapable of slowing the pace of evolution of immune evasion, and vaccinal protection against severe and fatal outcomes for COVID-19 patients is therefore not assured.