Multimodal artificial intelligence-based pathogenomics improves survival prediction in oral squamous cell carcinoma.

In this study, we aimed to develop a novel prognostic algorithm for oral squamous cell carcinoma (OSCC) using a combination of pathogenomics and AI-based techniques. We collected comprehensive clinical, genomic, and pathology data from a cohort of OSCC patients in the TCGA dataset and used machine learning and deep learning algorithms to identify relevant features that are predictive of survival outcomes. Our analyses included 406 OSCC patients. Initial analyses involved gene expression analyses, principal component analyses, gene enrichment analyses, and feature importance analyses. These insights were foundational for subsequent model development. Furthermore, we applied five machine learning/deep learning algorithms (Random Survival Forest, Gradient Boosting Survival Analysis, Cox PH, Fast Survival SVM, and DeepSurv) for survival prediction. Our initial analyses revealed relevant gene expression variations and biological pathways, laying the groundwork for robust feature selection in model building. The results showed that the multimodal model outperformed the unimodal models across all methods, with c-index values of 0.722 for RSF, 0.633 for GBSA, 0.625 for FastSVM, 0.633 for CoxPH, and 0.515 for DeepSurv. When considering only important features, the multimodal model continued to outperform the unimodal models, with c-index values of 0.834 for RSF, 0.747 for GBSA, 0.718 for FastSVM, 0.742 for CoxPH, and 0.635 for DeepSurv. Our results demonstrate the potential of pathogenomics and AI-based techniques in improving the accuracy of prognostic prediction in OSCC, which may ultimately aid in the development of personalized treatment strategies for patients with this devastating disease.

Synthetic 3D Spinal Vertebrae Reconstruction from Biplanar X-rays Utilizing Generative Adversarial Networks.

Computed tomography (CT) offers detailed insights into the internal anatomy of patients, particularly for spinal vertebrae examination. However, CT scans are associated with higher radiation exposure and cost compared to conventional X-ray imaging. In this study, we applied a Generative Adversarial Network (GAN) framework to reconstruct 3D spinal vertebrae structures from synthetic biplanar X-ray images, specifically focusing on anterior and lateral views. The synthetic X-ray images were generated using the DRRGenerator module in 3D Slicer by incorporating segmentations of spinal vertebrae in CT scans for the region of interest. This approach leverages a novel feature fusion technique based on X2CT-GAN to combine information from both views and employs a combination of mean squared error (MSE) loss and adversarial loss to train the generator, resulting in high-quality synthetic 3D spinal vertebrae CTs. A total of n = 440 CT data were processed. We evaluated the performance of our model using multiple metrics, including mean absolute error (MAE) (for each slice of the 3D volume (MAE0) and for the entire 3D volume (MAE)), cosine similarity, peak signal-to-noise ratio (PSNR), 3D peak signal-to-noise ratio (PSNR-3D), and structural similarity index (SSIM). The average PSNR was 28.394 dB, PSNR-3D was 27.432, SSIM was 0.468, cosine similarity was 0.484, MAE0 was 0.034, and MAE was 85.359. The results demonstrated the effectiveness of this approach in reconstructing 3D spinal vertebrae structures from biplanar X-rays, although some limitations in accurately capturing the fine bone structures and maintaining the precise morphology of the vertebrae were present. This technique has the potential to enhance the diagnostic capabilities of low-cost X-ray machines while reducing radiation exposure and cost associated with CT scans, paving the way for future applications in spinal imaging and diagnosis.

Artificial intelligence-based analysis of associations between learning curve and clinical outcomes in endoscopic and microsurgical lumbar decompression surgery.

PURPOSE: A common spine surgery procedure involves decompression of the lumbar spine. The impact of the surgeon's learning curve on relevant clinical outcomes is currently not well examined in the literature. A variety of machine learning algorithms have been investigated in this study to determine how a surgeon's learning curve and other clinical parameters will influence prolonged lengths of stay (LOS), extended operating times (OT), and complications, as well as whether these clinical parameters can be reliably predicted. METHODS: A retrospective monocentric cohort study of patients with lumbar spinal stenosis treated with microsurgical (MSD) and full-endoscopic (FED) decompression was conducted. The study included 206 patients with lumbar spinal stenosis who underwent FED (63; 30.6%) and MSD (118; 57.3%). Prolonged LOS and OT were defined as those exceeding the 75th percentile of the cohort. Furthermore, complications were assessed as a dependent variable. Using unsupervised learning, clusters were identified in the data, which helped distinguish between the early learning curve (ELC) and the late learning curve (LLC). From 15 algorithms, the top five algorithms that best fit the data were selected for each prediction task. We calculated the accuracy of prediction (Acc) and the area under the curve (AUC). The most significant predictors were determined using a feature importance analysis. RESULTS: For the FED group, the median number of surgeries with case surgery type at the time of surgery was 72 in the ELC group and 274 in the LLC group. FED patients did not significantly differ in outcome variables (LOS, OT, complication rate) between the ELC and LLC group. The random forest model demonstrated the highest mean accuracy and AUC across all folds for each classification task. For OT, it achieved an accuracy of 76.08% and an AUC of 0.89. For LOS, the model reached an accuracy of 83.83% and an AUC of 0.91. Lastly, in predicting complications, the random forest model attained the highest accuracy of 89.90% and an AUC of 0.94. Feature importance analysis indicated that LOS, OT, and complications were more significantly affected by patient characteristics than the surgical technique (FED versus MSD) or the surgeon's learning curve. CONCLUSIONS: A median of 72 cases of FED surgeries led to comparable clinical outcomes in the early learning curve phase compared to experienced surgeons. These outcomes seem to be more significantly affected by patient characteristics than the learning curve or the surgical technique. Several study variables, including the learning curve, can be used to predict whether lumbar decompression surgery will result in an increased LOS, OT, or complications. To introduce the provided prediction tools into clinics, the algorithms need to be implemented into open-source software and externally validated through large-scale randomized controlled trials.

Performance of artificial intelligence-based algorithms to predict prolonged length of stay after head and neck cancer surgery.

Background: Medical resource management can be improved by assessing the likelihood of prolonged length of stay (LOS) for head and neck cancer surgery patients. The objective of this study was to develop predictive models that could be used to determine whether a patient's LOS after cancer surgery falls within the normal range of the cohort. Methods: We conducted a retrospective analysis of a dataset consisting of 300 consecutive patients who underwent head and neck cancer surgery between 2017 and 2022 at a single university medical center. Prolonged LOS was defined as LOS exceeding the 75th percentile of the cohort. Feature importance analysis was performed to evaluate the most important predictors for prolonged LOS. We then constructed 7 machine learning and deep learning algorithms for the prediction modeling of prolonged LOS. Results: The algorithms reached accuracy values of 75.40 (radial basis function neural network) to 97.92 (Random Trees) for the training set and 64.90 (multilayer perceptron neural network) to 84.14 (Random Trees) for the testing set. The leading parameters predicting prolonged LOS were operation time, ischemia time, the graft used, the ASA score, the intensive care stay, and the pathological stages. The results revealed that patients who had a higher number of harvested lymph nodes (LN) had a lower probability of recurrence but also a greater LOS. However, patients with prolonged LOS were also at greater risk of recurrence, particularly when fewer (LN) were extracted. Further, LOS was more strongly correlated with the overall number of extracted lymph nodes than with the number of positive lymph nodes or the ratio of positive to overall extracted lymph nodes, indicating that particularly unnecessary lymph node extraction might be associated with prolonged LOS. Conclusions: The results emphasize the need for a closer follow-up of patients who experience prolonged LOS. Prospective trials are warranted to validate the present results.

Finetuning of GLIDE stable diffusion model for AI-based text-conditional image synthesis of dermoscopic images.

Background: The development of artificial intelligence (AI)-based algorithms and advances in medical domains rely on large datasets. A recent advancement in text-to-image generative AI is GLIDE (Guided Language to Image Diffusion for Generation and Editing). There are a number of representations available in the GLIDE model, but it has not been refined for medical applications. Methods: For text-conditional image synthesis with classifier-free guidance, we have fine-tuned GLIDE using 10,015 dermoscopic images of seven diagnostic entities, including melanoma and melanocytic nevi. Photorealistic synthetic samples of each diagnostic entity were created by the algorithm. Following this, an experienced dermatologist reviewed 140 images (20 of each entity), with 10 samples originating from artificial intelligence and 10 from original images from the dataset. The dermatologist classified the provided images according to the seven diagnostic entities. Additionally, the dermatologist was asked to indicate whether or not a particular image was created by AI. Further, we trained a deep learning model to compare the diagnostic results of dermatologist versus machine for entity classification. Results: The results indicate that the generated images possess varying degrees of quality and realism, with melanocytic nevi and melanoma having higher similarity to real images than other classes. The integration of synthetic images improved the classification performance of the model, resulting in higher accuracy and precision. The AI assessment showed superior classification performance compared to dermatologist. Conclusion: Overall, the results highlight the potential of synthetic images for training and improving AI models in dermatology to overcome data scarcity.

Clinical and radiomics feature-based outcome analysis in lumbar disc herniation surgery.

BACKGROUND: Low back pain is a widely prevalent symptom and the foremost cause of disability on a global scale. Although various degenerative imaging findings observed on magnetic resonance imaging (MRI) have been linked to low back pain and disc herniation, none of them can be considered pathognomonic for this condition, given the high prevalence of abnormal findings in asymptomatic individuals. Nevertheless, there is a lack of knowledge regarding whether radiomics features in MRI images combined with clinical features can be useful for prediction modeling of treatment success. The objective of this study was to explore the potential of radiomics feature analysis combined with clinical features and artificial intelligence-based techniques (machine learning/deep learning) in identifying MRI predictors for the prediction of outcomes after lumbar disc herniation surgery. METHODS: We included n = 172 patients who underwent discectomy due to disc herniation with preoperative T2-weighted MRI examinations. Extracted clinical features included sex, age, alcohol and nicotine consumption, insurance type, hospital length of stay (LOS), complications, operation time, ASA score, preoperative CRP, surgical technique (microsurgical versus full-endoscopic), and information regarding the experience of the performing surgeon (years of experience with the surgical technique and the number of surgeries performed at the time of surgery). The present study employed a semiautomatic region-growing volumetric segmentation algorithm to segment herniated discs. In addition, 3D-radiomics features, which characterize phenotypic differences based on intensity, shape, and texture, were extracted from the computed magnetic resonance imaging (MRI) images. Selected features identified by feature importance analyses were utilized for both machine learning and deep learning models (n = 17 models). RESULTS: The mean accuracy over all models for training and testing in the combined feature set was 93.31 ± 4.96 and 88.17 ± 2.58. The mean accuracy for training and testing in the clinical feature set was 91.28 ± 4.56 and 87.69 ± 3.62. CONCLUSIONS: Our results suggest a minimal but detectable improvement in predictive tasks when radiomics features are included. However, the extent of this advantage should be considered with caution, emphasizing the potential of exploring multimodal data inputs in future predictive modeling.

Automated Detection and Measurement of Dural Sack Cross-Sectional Area in Lumbar Spine MRI Using Deep Learning.

Lumbar spine magnetic resonance imaging (MRI) is a critical diagnostic tool for the assessment of various spinal pathologies, including degenerative disc disease, spinal stenosis, and spondylolisthesis. The accurate identification and quantification of the dural sack cross-sectional area are essential for the evaluation of these conditions. Current manual measurement methods are time-consuming and prone to inter-observer variability. Our study developed and validated deep learning models, specifically U-Net, Attention U-Net, and MultiResUNet, for the automated detection and measurement of the dural sack area in lumbar spine MRI, using a dataset of 515 patients with symptomatic back pain and externally validating the results based on 50 patient scans. The U-Net model achieved an accuracy of 0.9990 and 0.9987 on the initial and external validation datasets, respectively. The Attention U-Net model reported an accuracy of 0.9992 and 0.9989, while the MultiResUNet model displayed a remarkable accuracy of 0.9996 and 0.9995, respectively. All models showed promising precision, recall, and F1-score metrics, along with reduced mean absolute errors compared to the ground truth manual method. In conclusion, our study demonstrates the potential of these deep learning models for the automated detection and measurement of the dural sack cross-sectional area in lumbar spine MRI. The proposed models achieve high-performance metrics in both the initial and external validation datasets, indicating their potential utility as valuable clinical tools for the evaluation of lumbar spine pathologies. Future studies with larger sample sizes and multicenter data are warranted to validate the generalizability of the model further and to explore the potential integration of this approach into routine clinical practice.

Clinical and Radiological Outcome of Disc Arthroplasty for the Treatment of Cervical Spondylotic Myelopathy.

INTRODUCTION: The aim of this study is to evaluate the clinical and radiological results of cervical disc arthroplasty (CDA) in patients with cervical spondylotic myelopathy (CSM) using the CP ESP® disc prosthesis. MATERIALS AND METHODS: Prospectively collected data of 56 patients with CSM have been analyzed. The mean age at surgery was 35.6 years (range: 25-43 years). The mean follow-up was 28.2 months (range: 13-42 months). The range of motion (ROM) of the index segments, as well as upper and lower adjacent segments, was measured before surgery and at final follow-up. The C2-C7 sagittal vertical axis (SVA), C2-C7 cervical lordosis (CL), and T1 slope minus cervical lordosis (T1s-CL) were analyzed as well. Pain intensity was measured preoperatively and during follow-up using an 11-point numeric rating scale (NRS). Modified Japanese Orthopaedic Association (mJOA) score was assessed preoperatively and during follow-up for the clinical assessment of myelopathy. Surgical and implant-associated complications were analyzed as well. RESULTS: The NRS pain score improved from a mean of 7.4 (±1.1) preoperatively to a mean of 1.5 (±0.7) at last follow-up (p < 0.001). The mJOA score improved from a mean of 13.1 (±2.8) preoperatively to a mean of 14.8 (±2.3) at last follow-up (p < 0.001). The mean ROM of the index levels increased from 5.2° (±3.0) preoperatively to 7.3° (±3.2) at last follow-up (p < 0.05). Four patients developed heterotopic ossifications during follow-up. One patient developed permanent dysphonia. CONCLUSIONS: CDA showed good clinical and radiological outcome in this cohort of young patients. The motion of index segments could be preserved. CDA may be a viable treatment option in selected patients with CSM.

Celecoxib alleviates nociceptor sensitization mediated by interleukin-1beta-primed annulus fibrosus cells.

PURPOSE: This study aims to analyze the effect of pro-inflammatory cytokine-stimulated human annulus fibrosus cells (hAFCs) on the sensitization of dorsal root ganglion (DRG) cells. We further hypothesized that celecoxib (cxb) could inhibit hAFCs-induced DRG sensitization. METHODS: hAFCs from spinal trauma patients were stimulated with TNF-α or IL-1ß. Cxb was added on day 2. On day 4, the expression of pro-inflammatory and neurotrophic genes was evaluated using RT-qPCR. Levels of prostaglandin E2 (PGE-2), IL-8, and IL-6 were measured in the conditioned medium (CM) using ELISA. hAFCs CM was then applied to stimulate the DRG cell line (ND7/23) for 6 days. Then, calcium imaging (Fluo4) was performed to evaluate DRG cell sensitization. Both spontaneous and bradykinin-stimulated (0.5 µM) calcium responses were analyzed. The effects on primary bovine DRG cell culture were performed in parallel to the DRG cell line model. RESULTS: IL-1ß stimulation significantly enhanced the release of PGE-2 in hAFCs CM, while this increase was completely suppressed by 10 µM cxb. hAFCs revealed elevated IL-6 and IL-8 release following TNF-α and IL-1ß treatment, though cxb did not alter this. The effect of hAFCs CM on DRG cell sensitization was influenced by adding cxb to hAFCs; both the DRG cell line and primary bovine DRG nociceptors showed a lower sensitivity to bradykinin stimulation. CONCLUSION: Cxb can inhibit PGE-2 production in hAFCs in an IL-1ß-induced pro-inflammatory in vitro environment. The cxb applied to the hAFCs also reduces the sensitization of DRG nociceptors that are stimulated by the hAFCs CM.

Does the Pathologic Fracture Predict Severe Paralysis in Patients with Metastatic Epidural Spinal Cord Compression (MESCC)?-A Retrospective, Single-Center Cohort Analysis.

Currently, there is uncertainty about the predictive factors for metastatic epidural spinal cord compression (MESCC) and consecutive symptomatology in tumor patients. Prognostic algorithms for identifying patients at risk for paralysis are missing. The influence of the pathologic fracture on the patient's symptoms is widely discussed in the literature and we hypothesize that pathologic fractures contribute to spinal cord compression and are therefore predictive of severe paralysis. We tested this hypothesis in 136 patients who underwent surgery for spinal metastases. The most common primary cancers were prostate (24.3%, n = 33), breast (11.0%, n = 15), lung (10.3%, n = 14), and cancer of unknown primary (10.3%, n = 14). MESCC primarily affected the thoracic (77.2%, n = 105), followed by the lumbar (13.2%, n = 18) and cervical (9.6%, n = 13) spine. Pathologic fractures occurred in 63.2% (n = 86) of patients, mainly in osteolytic metastases. On the American spinal injury association (ASIA) impairment scale (AIS), 63.2% (n = 86) of patients exhibited AIS grade D and 36.8% (n = 50) AIS grade C-A preoperatively. The presence of a pathologic fracture alone did not predict severe paralysis (AIS C-A, p = 0.583). However, the duration of sensorimotor impairments, patient age, spinal instability neoplastic score (SINS), and the epidural spinal cord compression (ESCC) grade together predicted severe paralysis (p = 0.006) as did the ESCC grade 3 alone (p = 0.028). This is in contrast to previous studies that stated no correlation between the degree of spinal cord compression and the severity of neurologic impairments. Furthermore, the high percentage of pathologic fractures found in this study is above previously reported incidences. The risk factors identified can help to predict the development of paralysis and assist in the improvement of follow-up algorithms and the timing of therapeutic interventions.

Case-matched radiological and clinical outcome evaluation of interlaminar versus microsurgical decompression of lumbar spinal stenosis.

PURPOSE: Endoscopic spine surgery is a globally expanding technique advocated as less invasive for spinal stenosis treatment compared to the microsurgical approach. However, evidence on the efficiency of interlaminar full-endoscopic decompression (FED) vs. conventional microsurgical decompression (MSD) in patients with lumbar spinal stenosis is still scarce. We conducted a case-matched comparison for treatment success with consideration of clinical, laboratory, and radiologic predictors. METHODS: We included 88 consecutive patients (FED: 36/88, 40.9%; MSD: 52/88, 59.1%) presenting with lumbar central spinal stenosis. Surgery-related (operation time, complications, length of stay (LOS), American Society of Anesthesiologists physical status (ASA) score, C-reactive protein (CRP), white blood cell count, side of approach (unilateral/bilateral), patient-related outcome measures (PROMs) (Oswestry disability index (ODI), numeric rating scale of pain (NRS; leg-, back pain), EuroQol questionnaire (eQ-5D), core outcome measures index (COMI)), and radiological (dural sack cross-sectional area, Schizas score (SC), left and right lateral recess heights, and facet angles, respectively) parameters were extracted at different time points up to 1-year follow-up. The relationship of PROMs was analyzed using Spearman's rank correlation. Surgery-related outcome parameters were correlated with patient-centered and radiological outcomes utilizing a regression model to determine predictors for propensity score matching. RESULTS: Complication (most often residual sensorimotor deficits and restenosis due to hematoma) rates were higher in the FED (33.3%) than MSD (13.5%) group (p < 0.05), while all complications in the FED group were observed within the first 20 FED patients. Operation time was higher in the FED, whereas LOS was higher in the MSD group. Age, SC, CRP revealed significant associations with PROMs. We did not observe significant differences in the endoscopic vs. microsurgical group in PROMs. The correlation between ODI and COMI was significantly high, and both were inversely correlated with eQ-5D, whereas the correlations of these PROMs with NRS findings were less pronounced. CONCLUSIONS: Endoscopic treatment of lumbar spinal stenosis was similarly successful as the conventional microsurgical approach. Although FED was associated with higher complication rates in our single-center study experience, the distribution of complications indicated surgical learning curves to be the main factor of these findings. Future long-term prospective studies considering the surgical learning curve are warranted for reliable comparisons of these techniques.

Full-endoscopic versus conventional microsurgical therapy of lumbar disc herniation: a prospective, controlled, single-center, comprehensive cohort trial (FEMT-LDH trial).

BACKGROUND: Lumbar disc herniation is one of the leading causes of chronic low back pain. Surgery remains the therapy of choice when conservative approaches fail. Full-endoscopic approaches represent a promising alternative to the well-established microsurgical technique. However, high-grade evidence comparing these techniques is still scarce. METHODS: Patients presenting with lumbar disc herniation will be included. The intervention group will obtain full-endoscopic disc decompression, whereas the control group will be treated by microsurgical disc decompression. We will apply a comprehensive cohort study design involving a randomized and a prospective non-randomized study arm. Patients who do not consent to be randomized will be assigned to the non-randomized arm. The primary outcome will be the Oswestry Disability Index (ODI). Secondary outcomes involve the visual analog scale (VAS) of pain and the SF-36 health questionnaire. Furthermore, clinical characteristics including duration of hospital stay, operation time, and complications as well as laboratory markers, such as C-reactive protein, white blood cell counts, and interleukin 6 will be determined and compared. DISCUSSION: This study will significantly contribute to the current evidence available in the literature by evaluating the outcome of the full-endoscopic technique against the gold standard for lumbar disc herniation in a clinically relevant study setup. Additionally, the study design allows us to include patients not willing to be randomized in a prospective parallel study arm and to evaluate the impact of randomization on outcomes and include. The results could help to improve the future therapy in patients suffering from lumbar disc herniation. TRIAL REGISTRATION: This study was prospectively registered in The German Clinical Trials Register (DRKS), a German WHO primary registry, under the registration number: DRKS00025786. Registered on July 7, 2021.

In Vitro Characterization of a Tissue Renin-Angiotensin System in Human Nucleus Pulposus Cells.

Low back pain is a clinically highly relevant musculoskeletal burden and is associated with inflammatory as well as degenerative processes of the intervertebral disc. However, the pathophysiology and cellular pathways contributing to this devastating condition are still poorly understood. Based on previous evidence, we hypothesize that tissue renin-angiotensin system (tRAS) components, including the SARS-CoV-2 entry receptor angiotensin-converting enzyme 2 (ACE2), are present in human nucleus pulposus (NP) cells and associated with inflammatory and degenerative processes. Experiments were performed with NP cells from four human donors. The existence of angiotensin II, angiotensin II type 1 receptor (AGTR1), AGTR2, MAS-receptor (MasR), and ACE2 in human NP cells was validated with immunofluorescent staining and gene expression analysis. Hereafter, the cell viability was assessed after adding agonists and antagonists of the target receptors as well as angiotensin II in different concentrations for up to 48 h of exposure. A TNF-α-induced inflammatory in vitro model was employed to assess the impact of angiotensin II addition and the stimulation or inhibition of the tRAS receptors on inflammation, tissue remodeling, expression of tRAS markers, and the release of nitric oxide (NO) into the medium. Furthermore, protein levels of IL-6, IL-8, IL-10, and intracellular as well as secreted angiotensin II were assessed after exposing the cells to the substances, and inducible nitric oxide synthase (iNOS) levels were evaluated by utilizing Western blot. The existence of tRAS receptors and angiotensin II were validated in human NP cells. The addition of angiotensin II only showed a mild impact on gene expression markers. However, there was a significant increase in NO secreted by the cells. The gene expression ratios of pro-inflammatory/anti-inflammatory cytokines IL-6/IL-10, IL-8/IL-10, and TNF-α/IL-10 were positively correlated with the AGTR1/AGTR2 and AGTR1/MAS1 ratios, respectively. The stimulation of the AGTR2 MAS-receptor and the inhibition of the AGTR1 receptor revealed beneficial effects on the gene expression of inflammatory and tissue remodeling markers. This finding was also present at the protein level. The current data showed that tRAS components are expressed in human NP cells and are associated with inflammatory and degenerative processes. Further characterization of the associated pathways is warranted. The findings indicate that tRAS modulation might be a novel therapeutic approach to intervertebral disc disease.

Performance of Artificial Intelligence-Based Algorithms to Predict Prolonged Length of Stay after Lumbar Decompression Surgery.

BACKGROUND: Decompression of the lumbar spine is one of the most common procedures performed in spine surgery. Hospital length of stay (LOS) is a clinically relevant metric used to assess surgical success, patient outcomes, and socioeconomic impact. This study aimed to investigate a variety of machine learning and deep learning algorithms to reliably predict whether a patient undergoing decompression of lumbar spinal stenosis will experience a prolonged LOS. METHODS: Patients undergoing treatment for lumbar spinal stenosis with microsurgical and full-endoscopic decompression were selected within this retrospective monocentric cohort study. Prolonged LOS was defined as an LOS greater than or equal to the 75th percentile of the cohort (normal versus prolonged stay; binary classification task). Unsupervised learning with K-means clustering was used to find clusters in the data. Hospital stay classes were predicted with logistic regression, RandomForest classifier, stochastic gradient descent (SGD) classifier, K-nearest neighbors, Decision Tree classifier, Gaussian Naive Bayes (GaussianNB), support vector machines (SVM), a custom-made convolutional neural network (CNN), multilayer perceptron artificial neural network (MLP), and radial basis function neural network (RBNN) in Python. Prediction accuracy and area under the curve (AUC) were calculated. Feature importance analysis was utilized to find the most important predictors. Further, we developed a decision tree based on the Chi-square automatic interaction detection (CHAID) algorithm to investigate cut-offs of predictors for clinical decision-making. RESULTS: 236 patients and 14 feature variables were included. K-means clustering separated data into two clusters distinguishing the data into two patient risk characteristic groups. The algorithms reached AUCs between 67.5% and 87.3% for the classification of LOS classes. Feature importance analysis of deep learning algorithms indicated that operation time was the most important feature in predicting LOS. A decision tree based on CHAID could predict 84.7% of the cases. CONCLUSIONS: Machine learning and deep learning algorithms can predict whether patients will experience an increased LOS following lumbar decompression surgery. Therefore, medical resources can be more appropriately allocated to patients who are at risk of prolonged LOS.

One-Year Clinical Outcomes of Minimal-Invasive Dorsal Percutaneous Fixation of Thoracolumbar Spine Fractures.

Introduction: Minimal-invasive instrumentation techniques have become a workhorse in spine surgery and require constant clinical evaluations. We sought to analyze patient-reported outcome measures (PROMs) and clinicopathological characteristics of thoracolumbar fracture stabilizations utilizing a minimal-invasive percutaneous dorsal screw-rod system. Methods: We included all patients with thoracolumbar spine fractures who underwent minimal-invasive percutaneous spine stabilization in our clinics since inception and who have at least 1 year of follow-up data. Clinical characteristics (length of hospital stay (LOS), operation time (OT), and complications), PROMs (preoperative (pre-op), 3-weeks postoperative (post-op), 1-year postoperative: eq5D, COMI, ODI, NRS back pain), and laboratory markers (leucocytes, c-reactive protein (CRP)) were analyzed, finding significant associations between these study variables and PROMs. Results: A total of 68 patients (m: 45.6%; f: 54.4%; mean age: 76.9 ± 13.9) were included. The most common fracture types according to the AO classification were A3 (40.3%) and A4 (40.3%), followed by B2 (7.46%) and B1 (5.97%). The Median American Society of Anesthesiologists (ASA) score was 3 (range: 1−4). Stabilized levels ranged from TH4 to L5 (mean number of targeted levels: 4.25 ± 1.4), with TH10-L2 (12/68) and TH11-L3 (11/68) being the most frequent site of surgery. Mean OT and LOS were 92.2 ± 28.2 min and 14.3 ± 6.9 days, respectively. We observed 9/68 complications (13.2%), mostly involving screw misalignments and loosening. CRP increased from 24.9 ± 33.3 pre-op to 34.8 ± 29.9 post-op (p < 0.001), whereas leucocyte counts remained stable. All PROMs showed a marked significant improvement for both 3-week and 1-year evaluations compared to the preoperative situation. Interestingly, we did not find an impact of OT, LOS, lab markers, complications, and other clinical characteristics on PROMs. Notably, a higher number of stabilized levels did not affect PROMs. Conclusions: Minimal-invasive stabilization of thoracolumbar fractures utilizing a dorsal percutaneous approach resulted in significant PROM outcome improvements, although we observed a complication rate of 13.2% for up to 1 year of follow-up. PROMs were not significantly associated with clinicopathological characteristics, technique-related variables, or the number of targeted levels.

Artificial Intelligence-Driven Prediction Modeling and Decision Making in Spine Surgery Using Hybrid Machine Learning Models.

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.

A Proinflammatory, Degenerative Organ Culture Model to Simulate Early-Stage Intervertebral Disc Disease.

Symptomatic intervertebral disc (IVD) degeneration (IDD) is a major socioeconomic burden and is characterized by inflammation and tissue degradation. Due to the lack of causative therapies, there is an urgent need for innovative experimental organ culture models to study the mechanisms involved in the progression of the disease, find therapeutic targets, and reduce the need for animal models. We here present a novel, three-dimensional organ culture model protocol mimicking the proinflammatory and catabolic microenvironment, which is present during IDD. Initially, bovine caudal IVDs were dissected, cleaned, and cultured in the tissue culture medium. Dynamic physiologic or pathologic loading was applied in a custom-made bioreactor for 2 hours per day. IVDs were assigned to a control group (high glucose medium, physiological loading, phosphate-buffered saline injection) and a pathological group (low glucose medium, pathological loading, tumor necrosis factor-alpha injection) for four days. Gene expression analysis from collected nucleus pulposus cells of the IVDs and enzyme-linked immunosorbent assay of the conditioned organ culture media was performed. Our data revealed a higher expression of inflammatory markers and reduced disc heights after loading in the pathological group compared to the control group. This protocol is reliable to simulate IVD inflammation and degeneration and can be further expanded to broaden its application scope.

Corrective Osteotomy of Upper Extremity Malunions Using Three-Dimensional Planning and Patient-Specific Surgical Guides: Recent Advances and Perspectives.

Malunions of the upper extremity can result in severe functional problems and increase the risk of osteoarthritis. The surgical reconstruction of complex malunions can be technically challenging. Recent advances in computer-assisted orthopedic surgery provide an innovative solution for complex three-dimensional (3-D) reconstructions. This study aims to evaluate the clinical applicability of 3-D computer-assisted planning and surgery for upper extremity malunions. Hence, we provide a summary of evidence on this topic and highlight recent advances in this field. Further, we provide a practical implementation of this therapeutic approach based on three cases of malunited forearm fractures treated with corrective osteotomy using preoperative three-dimensional simulation and patient-specific surgical guides. All three cases, one female (56 years old) and two males (18 and 26 years old), had painful restrictions in range of motion (ROM) due to forearm malunions and took part in clinical and radiologic assessments. Postoperative evaluation of patient outcomes showed a substantial increase in range of motion, reduction of preoperatively reported pain, and an overall improvement of patients' satisfaction. The therapeutic approach used in these cases resulted in an excellent anatomical and functional reconstruction and was assessed as precise, safe, and reliable. Based on current evidence and our results, the 3-D preoperative planning technique could be the new gold standard in the treatment of complex upper extremity malunions in the future.

Clinicopathologic characteristics, metastasis-free survival, and skeletal-related events in 628 patients with skeletal metastases in a tertiary orthopedic and trauma center.

BACKGROUND: Skeletal-related events (SREs) due to bone metastases (BM) significantly impact the morbidity and mortality of cancer patients. The present study sought to investigate clinicopathological characteristics, metastasis-free survival (MFS), and SREs in patients referred to a tertiary orthopedic and trauma center. METHODS: Data were retrieved from electronic health records (n=628). Survival curves were estimated utilizing the Kaplan-Meier method. The Cox regression model was used to determine factors influencing MFS based on estimated hazard ratios (HRs). RESULTS: Breast (55.8%) and lung (18.2%), and lung (32.9%) and prostate (16.8%) cancer were the most common cancer types in our cohort in women and men, respectively. Fifteen percent of patients presented with BM as the first manifestation of tumor disease, 23% had metastasis diagnosis on the same day of primary tumor diagnosis or within 3 months, and 62% developed BM at least 3 months after primary tumor diagnosis. Osteolytic BM were predominant (72.3%) and most commonly affecting the spine (23%). Overall median MFS was 45 months (32 (men) vs. 53 (women) months). MFS was shortest in the lung (median 15 months, 95% CI 8.05-19) and longest in breast cancer (median 82 months, 95% CI 65.29-94). Age (≥ 60 vs. < 60 years) and primary cancer grading of ≥2 vs. 1 revealed prognostic relevance. CONCLUSION: Women with breast or lung cancer, men with lung or prostate cancer, age ≥60 years, male sex, and primary cancer grading ≥2 are associated with increased risk for MBD. Intensified follow-up programs may reduce the risk of SREs and associated morbidity and mortality.

Proinflammatory intervertebral disc cell and organ culture models induced by tumor necrosis factor alpha.

Inflammation plays an important role in the pathogenesis of intervertebral disc (IVD) degeneration. The proinflammatory cytokine tumor necrosis factor alpha (TNF-α) has shown markedly higher expression in degenerated human disc tissue compared with healthy controls. Anti-inflammatory treatment targeting TNF-α has shown to alleviate discogenic pain in patients with low back pain. Therefore, in vitro and ex vivo inflammatory models utilizing TNF-α provide relevant experimental conditions for drug development in disc degeneration research. The current method article addressed several specific questions related to the model establishment. (a) The effects of bovine and human recombinant TNF-α on bovine nucleus pulposus (NP) cells were compared. (b) The required dose for an inflammatory IVD organ culture model with intradiscal TNF-α injection was studied. (c) The effect of TNF-α blocking at different stages of inflammation was evaluated. Outcomes revealed that bovine and human recombinant TNF-α induced equivalent inflammatory effects in bovine NP cells. A bovine whole IVD inflammatory model was established by intradiscal injection of 100 ng TNF-α/ cm3 disc volume, as indicated by increased nitric oxide, glycosaminoglycan, interleukin 6 (IL-6), and interleukin 8 (IL-8) release in culture media, and upregulation of MMP3, ADAMTS4, IL-8, IL-6, and cyclooxygenase (COX)-2 expression in NP tissue. However, results in human NP cells showed that the time point of anti-inflammatory treatment was crucial to achieve significant effects. Furthermore, anticatabolic therapy in conjunction with TNF-α inhibition would be required to slow down the pathologic cascade of disc degeneration.