Healthcare systems worldwide generate vast amounts of data from many different sources. Although of high complexity for a human being, it is essential to determine the patterns and minor variations in the genomic, radiological, laboratory, or clinical data that reliably differentiate phenotypes or allow high predictive accuracy in health-related tasks. Convolutional neural networks (CNN) are increasingly applied to image data for various tasks. Its use for non-imaging data becomes feasible through different modern machine learning techniques, converting non-imaging data into images before inputting them into the CNN model. Considering also that healthcare providers do not solely use one data modality for their decisions, this approach opens the door for multi-input/mixed data models which use a combination of patient information, such as genomic, radiological, and clinical data, to train a hybrid deep learning model. Thus, this reflects the main characteristic of artificial intelligence: simulating natural human behavior. The present review focuses on key advances in machine and deep learning, allowing for multi-perspective pattern recognition across the entire information set of patients in spine surgery. This is the first review of artificial intelligence focusing on hybrid models for deep learning applications in spine surgery, to the best of our knowledge. This is especially interesting as future tools are unlikely to use solely one data modality. The techniques discussed could become important in establishing a new approach to decision-making in spine surgery based on three fundamental pillars: (1) patient-specific, (2) artificial intelligence-driven, (3) integrating multimodal data. The findings reveal promising research that already took place to develop multi-input mixed-data hybrid decision-supporting models. Their implementation in spine surgery may hence be only a matter of time.

Capillary leak syndrome (CLS) represents a phenotype of increased fluid extravasation, resulting in intravascular hypovolemia, extravascular edema formation and ultimately hypoperfusion. While endothelial permeability is an evolutionary preserved physiological process needed to sustain life, excessive fluid leak—often caused by systemic inflammation—can have detrimental effects on patients’ outcomes. This article delves into the current understanding of CLS pathophysiology, diagnosis and potential treatments. Systemic inflammation leading to a compromise of endothelial cell interactions through various signaling cues (e.g., the angiopoietin–Tie2 pathway), and shedding of the glycocalyx collectively contribute to the manifestation of CLS. Capillary permeability subsequently leads to the seepage of protein-rich fluid into the interstitial space. Recent insights into the importance of the sub-glycocalyx space and preserving lymphatic flow are highlighted for an in-depth understanding. While no established diagnostic criteria exist and CLS is frequently diagnosed by clinical characteristics only, we highlight more objective serological and (non)-invasive measurements that hint towards a CLS phenotype. While currently available treatment options are limited, we further review understanding of fluid resuscitation and experimental approaches to target endothelial permeability. Despite the improved understanding of CLS pathophysiology, efforts are needed to develop uniform diagnostic criteria, associate clinical consequences to these criteria, and delineate treatment options. Graphical Abstract

The success rate of dental implants depends on primary and secondary stability. We investigate predictive factors for future risk stratification models. We retrospectively analyze 272 patients with a total of 582 implants. Implant stability is measured with resonance frequency analysis and evaluated based on the implant stability quotient (ISQ). A linear regression model with regression coefficients (reg. coeff.) and its 95% confidence interval (95% CI) is applied to assess predictive factors for implant stability. Implant diameter (reg. coeff.: 3.28; 95% CI: 1.89–4.66, p < 0.001), implant length (reg. coeff.: 0.67, 95% CI: 0.26–1.08, p < 0.001), and implant localization (maxillary vs. mandibular, reg. coeff.: −7.45, 95% CI: −8.70–(−6.20), p < 0.001) are significant prognostic factors for primary implant stability. An increase in ISQ between insertion and exposure is significantly correlated with healing time (reg. coeff.: 0.11, 95% CI: 0.04–0.19). Patients with maxillary implants have lower ISQ at insertion but show a higher increase in ISQ after insertion than patients with mandibular implants. We observe positive associations between primary implant stability and implant diameter, implant length, and localization (mandibular vs. maxillary). An increase in implant stability between insertion and exposure is significantly correlated with healing time and is higher for maxillary implants. These predictive factors should be further evaluated in prospective cohort studies to develop future preoperative risk-stratification models.

Evidence has arisen in recent years suggesting that a tissue renin-angiotensin system (tRAS) is involved in the progression of various human diseases. This system contains two regulatory pathways: a pathological pro-inflammatory pathway containing the Angiotensin Converting Enzyme (ACE)/Angiotensin II (AngII)/Angiotensin II receptor type 1 (AGTR1) axis and a protective anti-inflammatory pathway involving the Angiotensin II receptor type 2 (AGTR2)/ACE2/Ang1–7/MasReceptor axis. Numerous studies reported the positive effects of pathologic tRAS pathway inhibition and protective tRAS pathway stimulation on the treatment of cardiovascular, inflammatory, and autoimmune disease and the progression of neuropathic pain. Cell senescence and aging are known to be related to RAS pathways. Further, this system directly interacts with SARS-CoV 2 and seems to be an important target of interest in the COVID-19 pandemic. This review focuses on the involvement of tRAS in the progression of the mentioned diseases from an interdisciplinary clinical perspective and highlights therapeutic strategies that might be of major clinical importance in the future.

Although CAD/CAM ceramics present a promising alternative to metal-ceramic fixed dental prostheses, little is known about their mid- and long-term clinical performance. This systematic review aims to estimate the survival and success rates and describes the underlying complication characteristics for CAD/CAM tooth-supported zirconia- and lithium disilicate-based fixed dental prostheses (FDPs). We systematically searched MEDLINE and Web of Science to find relevant prospective studies with a follow-up of at least one year. We estimated pooled 1-, 5-, and 10-year survival and success rates by combining the collected data in a Poisson regression model. Descriptive statistics were conducted to evaluate the distribution of failures and complications in the included studies. Risk of bias for the included studies was assessed with an adapted checklist for single-arm trials. Pooled estimated 1-, 5-, and 10-year survival rates ranged from 93.80% to 94.66%, 89.67% to 91.1%, and 79.33% to 82.20%, respectively. The corresponding success rates excluding failures, but including any other types of intervention were 94.53% to 96.77%, 90.89% to 94.62%, and 81.78% to 89.25%. Secondary caries was the most frequent cause of failure, followed by chipping of the veneering. The most common cause of complication excluding failures but requiring intervention was chipping of the veneering. Risk of bias was generally acceptable for the included studies, with seven studies associated with low risk of bias, eight studies with a moderate risk of bias, and three studies with serious risk of bias. The current meta-analysis on CAD/CAM-supported FDPs revealed satisfying survival and success rates for up to 10 years of exposure. More prospective studies focusing on long-term performance are needed to strengthen the evidence currently available in the literature.

Background: Ex vivo fluorescent confocal microscopy (FCM) is a novel and effective method for a fast-automatized histological tissue examination. In contrast, conventional diagnostic methods are primarily based on the skills of the histopathologist. In this study, we investigated the potential of convolutional neural networks (CNNs) for automatized classification of oral squamous cell carcinoma via ex vivo FCM imaging for the first time. Material and Methods: Tissue samples from 20 patients were collected, scanned with an ex vivo confocal microscope immediately after resection, and investigated histopathologically. A CNN architecture (MobileNet) was trained and tested for accuracy. Results: The model achieved a sensitivity of 0.47 and specificity of 0.96 in the automated classification of cancerous tissue in our study. Conclusion: In this preliminary work, we trained a CNN model on a limited number of ex vivo FCM images and obtained promising results in the automated classification of cancerous tissue. Further studies using large sample sizes are warranted to introduce this technology into clinics.