Machine learning-based preoperative analytics for the prediction of anastomotic leakage in colorectal surgery: a swiss pilot study.

BACKGROUND: Anastomotic leakage (AL), a severe complication following colorectal surgery, arises from defects at the anastomosis site. This study evaluates the feasibility of predicting AL using machine learning (ML) algorithms based on preoperative data. METHODS: We retrospectively analyzed data including 21 predictors from patients undergoing colorectal surgery with bowel anastomosis at four Swiss hospitals. Several ML algorithms were applied for binary classification into AL or non-AL groups, utilizing a five-fold cross-validation strategy with a 90% training and 10% validation split. Additionally, a holdout test set from an external hospital was employed to assess the models' robustness in external validation. RESULTS: Among 1244 patients, 112 (9.0%) suffered from AL. The Random Forest model showed an AUC-ROC of 0.78 (SD: ± 0.01) on the internal test set, which significantly decreased to 0.60 (SD: ± 0.05) on the external holdout test set comprising 198 patients, including 7 (3.5%) with AL. Conversely, the Logistic Regression model demonstrated more consistent AUC-ROC values of 0.69 (SD: ± 0.01) on the internal set and 0.61 (SD: ± 0.05) on the external set. Accuracy measures for Random Forest were 0.82 (SD: ± 0.04) internally and 0.87 (SD: ± 0.08) externally, while Logistic Regression achieved accuracies of 0.81 (SD: ± 0.10) and 0.88 (SD: ± 0.15). F1 Scores for Random Forest moved from 0.58 (SD: ± 0.03) internally to 0.51 (SD: ± 0.03) externally, with Logistic Regression maintaining more stable scores of 0.53 (SD: ± 0.04) and 0.51 (SD: ± 0.02). CONCLUSION: In this pilot study, we evaluated ML-based prediction models for AL post-colorectal surgery and identified ten patient-related risk factors associated with AL. Highlighting the need for multicenter data, external validation, and larger sample sizes, our findings emphasize the potential of ML in enhancing surgical outcomes and inform future development of a web-based application for broader clinical use.

A randomized controlled trial of social media promotion in neurosurgical publishing.

The importance of social media has seen a dramatic increase in recent times, but much about its influence in academia is still unknown. To date, no comparative studies analysing the effect of social media promotion on citation counts have been undertaken in neurosurgical publishing. We randomized 177 articles published in Acta Neurochirurgica from May to September 2020. The 89 articles in the intervention group received a standardized social media promotion through one post on our official Twitter/X account, whereas the 88 articles in the control group did not receive any social media promotion. Citation counts, website visits and PDF downloads were tracked at one and two years post-promotion. We found no significant difference in number of citations at one year post-promotion (Intervention: 1.85 ± 3.94 vs. Control: 2.67 ± 6.65, p = 0.322) or at two years (5.35 ± 7.39 vs. 7.09 ± 12.1, p = 0.249). Similarly, no difference was detected in website visits at one (587.46 ± 568.04 vs. 590.65 ± 636.25, p = 0.972) or two years (865.79 ± 855.80 vs. 896.31 ± 981.97, p = 0.826) and PDF downloads at one (183.40 ± 152.02 vs. 187.78 ± 199.01, p = 0.870) or two years (255.99 ± 218.97 vs. 260.97 ± 258.44, p = 0.890). In a randomized study, a structured promotion of general neurosurgical articles on Twitter/X did not significantly impact citation count, website visits, or PDF downloads compared to no social media promotion. Combined with published evidence to date, the impact of social media on citation counts in academic publishing ultimately remains unclear.

Identifying clusters of objective functional impairment in patients with degenerative lumbar spinal disease using unsupervised learning.

OBJECTIVES: The five-repetition sit-to-stand (5R-STS) test was designed to capture objective functional impairment (OFI), and thus provides an adjunctive dimension in patient assessment. It is conceivable that there are different subsets of patients with OFI and degenerative lumbar disease. We aim to identify clusters of objectively functionally impaired individuals based on 5R-STS and unsupervised machine learning (ML). METHODS: Data from two prospective cohort studies on patients with surgery for degenerative lumbar disease and 5R-STS times of ≥ 10.5 s-indicating presence of OFI. K-means clustering-an unsupervised ML algorithm-was applied to identify clusters of OFI. Cluster hallmarks were then identified using descriptive and inferential statistical analyses. RESULTS: We included 173 patients (mean age [standard deviation]: 46.7 [12.7] years, 45% male) and identified three types of OFI. OFI Type 1 (57 pts., 32.9%), Type 2 (81 pts., 46.8%), and Type 3 (35 pts., 20.2%) exhibited mean 5R-STS test times of 14.0 (3.2), 14.5 (3.3), and 27.1 (4.4) seconds, respectively. The grades of OFI according to the validated baseline severity stratification of the 5R-STS increased significantly with each OFI type, as did extreme anxiety and depression symptoms, issues with mobility and daily activities. Types 1 and 2 are characterized by mild to moderate OFI-with female gender, lower body mass index, and less smokers as Type I hallmarks. CONCLUSIONS: Unsupervised learning techniques identified three distinct clusters of patients with OFI that may represent a more holistic clinical classification of patients with OFI than test-time stratifications alone, by accounting for individual patient characteristics.

Calculation of the minimum clinically important difference (MCID) using different methodologies: case study and practical guide.

INTRODUCTION: Establishing thresholds of change that are actually meaningful for the patient in an outcome measurement instrument is paramount. This concept is called the minimum clinically important difference (MCID). We summarize available MCID calculation methods relevant to spine surgery, and outline key considerations, followed by a step-by-step working example of how MCID can be calculated, using publicly available data, to enable the readers to follow the calculations themselves. METHODS: Thirteen MCID calculations methods were summarized, including anchor-based methods, distribution-based methods, Reliable Change Index, 30% Reduction from Baseline, Social Comparison Approach and the Delphi method. All methods, except the latter two, were used to calculate MCID for improvement of Zurich Claudication Questionnaire (ZCQ) Symptom Severity of patients with lumbar spinal stenosis. Numeric Rating Scale for Leg Pain and Japanese Orthopaedic Association Back Pain Evaluation Questionnaire Walking Ability domain were used as anchors. RESULTS: The MCID for improvement of ZCQ Symptom Severity ranged from 0.8 to 5.1. On average, distribution-based methods yielded lower MCID values, than anchor-based methods. The percentage of patients who achieved the calculated MCID threshold ranged from 9.5% to 61.9%. CONCLUSIONS: MCID calculations are encouraged in spinal research to evaluate treatment success. Anchor-based methods, relying on scales assessing patient preferences, continue to be the "gold-standard" with receiver operating characteristic curve approach being optimal. In their absence, the minimum detectable change approach is acceptable. The provided explanation and step-by-step example of MCID calculations with statistical code and publicly available data can act as guidance in planning future MCID calculation studies.

Multicenter external validation of prediction models for clinical outcomes after spinal fusion for lumbar degenerative disease.

BACKGROUND: Clinical prediction models (CPM), such as the SCOAP-CERTAIN tool, can be utilized to enhance decision-making for lumbar spinal fusion surgery by providing quantitative estimates of outcomes, aiding surgeons in assessing potential benefits and risks for each individual patient. External validation is crucial in CPM to assess generalizability beyond the initial dataset. This ensures performance in diverse populations, reliability and real-world applicability of the results. Therefore, we externally validated the tool for predictability of improvement in oswestry disability index (ODI), back and leg pain (BP, LP). METHODS: Prospective and retrospective data from multicenter registry was obtained. As outcome measure minimum clinically important change was chosen for ODI with ≥ 15-point and ≥ 2-point reduction for numeric rating scales (NRS) for BP and LP 12 months after lumbar fusion for degenerative disease. We externally validate this tool by calculating discrimination and calibration metrics such as intercept, slope, Brier Score, expected/observed ratio, Hosmer-Lemeshow (HL), AUC, sensitivity and specificity. RESULTS: We included 1115 patients, average age 60.8 ± 12.5 years. For 12-month ODI, area-under-the-curve (AUC) was 0.70, the calibration intercept and slope were 1.01 and 0.84, respectively. For NRS BP, AUC was 0.72, with calibration intercept of 0.97 and slope of 0.87. For NRS LP, AUC was 0.70, with calibration intercept of 0.04 and slope of 0.72. Sensitivity ranged from 0.63 to 0.96, while specificity ranged from 0.15 to 0.68. Lack of fit was found for all three models based on HL testing. CONCLUSIONS: Utilizing data from a multinational registry, we externally validate the SCOAP-CERTAIN prediction tool. The model demonstrated fair discrimination and calibration of predicted probabilities, necessitating caution in applying it in clinical practice. We suggest that future CPMs focus on predicting longer-term prognosis for this patient population, emphasizing the significance of robust calibration and thorough reporting.

From molecular signatures to radiomics: tailoring neurooncological strategies through forecasting of glioma growth.

OBJECTIVE: Contemporary oncological paradigms for adjuvant treatment of low- and intermediate-grade gliomas are often guided by a limited array of parameters, overlooking the dynamic nature of the disease. The authors' aim was to develop a comprehensive multivariate glioma growth model based on multicentric data, to facilitate more individualized therapeutic strategies. METHODS: Random slope models with subject-specific random intercepts were fitted to a retrospective cohort of grade II and III gliomas from the database at Kepler University Hospital (n = 191) to predict future mean tumor diameters. Deep learning-based radiomics was used together with a comprehensive clinical dataset and evaluated on an external prospectively collected validation cohort from University Hospital Zurich (n = 9). Prediction quality was assessed via mean squared prediction error. RESULTS: A mean squared prediction error of 0.58 cm for the external validation cohort was achieved, indicating very good prognostic value. The mean ± SD time to adjuvant therapy was 28.7 ± 43.3 months and 16.1 ± 14.6 months for the training and validation cohort, respectively, with a mean of 6.2 ± 5 and 3.6 ± 0.7, respectively, for number of observations. The observed mean tumor diameter per year was 0.38 cm (95% CI 0.25-0.51) for the training cohort, and 1.02 cm (95% CI 0.78-2.82) for the validation cohort. Glioma of the superior frontal gyrus showed a higher rate of tumor growth than insular glioma. Oligodendroglioma showed less pronounced growth, anaplastic astrocytoma-unlike anaplastic oligodendroglioma-was associated with faster tumor growth. Unlike the impact of extent of resection, isocitrate dehydrogenase (IDH) had negligible influence on tumor growth. Inclusion of radiomics variables significantly enhanced the prediction performance of the random slope model used. CONCLUSIONS: The authors developed an advanced statistical model to predict tumor volumes both pre- and postoperatively, using comprehensive data prior to the initiation of adjuvant therapy. Using radiomics enhanced the precision of the prediction models. Whereas tumor extent of resection and topology emerged as influential factors in tumor growth, the IDH status did not. This study emphasizes the imperative of advanced computational methods in refining personalized low-grade glioma treatment, advocating a move beyond traditional paradigms.

A critical moment in machine learning in medicine: on reproducible and interpretable learning.

Over the past two decades, advances in computational power and data availability combined with increased accessibility to pre-trained models have led to an exponential rise in machine learning (ML) publications. While ML may have the potential to transform healthcare, this sharp increase in ML research output without focus on methodological rigor and standard reporting guidelines has fueled a reproducibility crisis. In addition, the rapidly growing complexity of these models compromises their interpretability, which currently impedes their successful and widespread clinical adoption. In medicine, where failure of such models may have severe implications for patients' health, the high requirements for accuracy, robustness, and interpretability confront ML researchers with a unique set of challenges. In this review, we discuss the semantics of reproducibility and interpretability, as well as related issues and challenges, and outline possible solutions to counteracting the "black box". To foster reproducibility, standard reporting guidelines need to be further developed and data or code sharing encouraged. Editors and reviewers may equally play a critical role by establishing high methodological standards and thus preventing the dissemination of low-quality ML publications. To foster interpretable learning, the use of simpler models more suitable for medical data can inform the clinician how results are generated based on input data. Model-agnostic explanation tools, sensitivity analysis, and hidden layer representations constitute further promising approaches to increase interpretability. Balancing model performance and interpretability are important to ensure clinical applicability. We have now reached a critical moment for ML in medicine, where addressing these issues and implementing appropriate solutions will be vital for the future evolution of the field.

Early versus delayed mobilization after aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis of efficacy and safety.

OBJECTIVE: A central tenet of Enhanced Recovery After Surgery (ERAS) is evidence-based medicine. Survivors of aneurysmal subarachnoid hemorrhage (aSAH) constitute a fragile patient population prone to prolonged hospitalization within neurointensive care units (NICUs), prolonged immobilization, and a range of nosocomial adverse events. Potentially, well-monitored early mobilization (EM) could constitute a beneficial element of ERAS protocols in this population. Therefore, the objective was to summarize the available evidence on EM strategies in patients with aSAH. METHODS: The authors retrieved prospective and retrospective studies that reported efficacy or safety data on EM (defined as EM in the NICU starting ≤ 7 days after ictus) versus delayed mobilization (DM) (any strategy that comparatively delayed mobilization) after aSAH and were published after January 1, 2000, in PubMed/MEDLINE, Embase, and the Cochrane Library. Random-effects meta-analysis was performed. RESULTS: Ten studies analyzing 1292 patients were included for quantitative synthesis, including 1 randomized, 1 prospective nonrandomized, and 8 retrospective studies. Modified Rankin Scale scores at discharge were not different between the EM and DM groups (mean difference [MD] [95% CI] -0.86 [-2.93 to 1.20] points, p = 0.41). Hospital length of stay in days was markedly reduced in the EM group (MD [95% CI] -6.56 [-10.64 to -2.47] days, p = 0.002). Although there was a statistically significant reduction in radiological vasospasms (OR [95% CI] 0.65 [0.44-0.97], p = 0.03), the reduction in clinically relevant vasospasms was nonsignificant (OR [95% CI] 0.63 [0.31-1.26], p = 0.19). The odds of shunting were significantly lower in the EM group (OR [95% CI] 0.61 [0.39-0.95], p = 0.03). The rates of mortality, pneumonia, and thrombosis were similar among groups (p > 0.05). CONCLUSIONS: Due to a lack of high-quality studies, vastly varying protocols, and resulting statistical clinical and statistical heterogeneity, the level of evidence for recommendations regarding EM in patients with aSAH remains low. The currently available data indicated that mobilization within the first 5 days after aneurysm repair was feasible and safe without significant excessive adverse events, that neurological outcome with EM was almost certainly not worse than with prolonged immobilization, and that there was likely at least some reduction in length of hospital stay. Radiological and clinical vasospasms were not more frequent-with signals even trending toward a decrease-in patients who mobilized early. Higher-quality studies and implementation of full ERAS protocols are necessary to evaluate efficacy and safety with a higher level of evidence and to guide practical implementation through increased standardization. Clinical trial registration no.: CRD42023432828 (www.crd.york.ac.uk/prospero).

Whole Spine Segmentation Using Object Detection and Semantic Segmentation.

OBJECTIVE: Virtual and augmented reality have enjoyed increased attention in spine surgery. Preoperative planning, pedicle screw placement, and surgical training are among the most studied use cases. Identifying osseous structures is a key aspect of navigating a 3-dimensional virtual reconstruction. To automate the otherwise time-consuming process of labeling vertebrae on each slice individually, we propose a fully automated pipeline that automates segmentation on computed tomography (CT) and which can form the basis for further virtual or augmented reality application and radiomic analysis. METHODS: Based on a large public dataset of annotated vertebral CT scans, we first trained a YOLOv8m (You-Only-Look-Once algorithm, Version 8 and size medium) to detect each vertebra individually. On the then cropped images, a 2D-U-Net was developed and externally validated on 2 different public datasets. RESULTS: Two hundred fourteen CT scans (cervical, thoracic, or lumbar spine) were used for model training, and 40 scans were used for external validation. Vertebra recognition achieved a mAP50 (mean average precision with Jaccard threshold of 0.5) of over 0.84, and the segmentation algorithm attained a mean Dice score of 0.75 ± 0.14 at internal, 0.77 ± 0.12 and 0.82 ± 0.14 at external validation, respectively. CONCLUSION: We propose a 2-stage approach consisting of single vertebra labeling by an object detection algorithm followed by semantic segmentation. In our externally validated pilot study, we demonstrate robust performance for our object detection network in identifying individual vertebrae, as well as for our segmentation model in precisely delineating the bony structures.

TomoRay: Generating Synthetic Computed Tomography of the Spine From Biplanar Radiographs.

OBJECTIVE: Computed tomography (CT) imaging is a cornerstone in the assessment of patients with spinal trauma and in the planning of spinal interventions. However, CT studies are associated with logistical problems, acquisition costs, and radiation exposure. In this proof-of-concept study, the feasibility of generating synthetic spinal CT images using biplanar radiographs was explored. This could expand the potential applications of x-ray machines pre-, post-, and even intraoperatively. METHODS: A cohort of 209 patients who underwent spinal CT imaging from the VerSe2020 dataset was used to train the algorithm. The model was subsequently evaluated using an internal and external validation set containing 55 from the VerSe2020 dataset and a subset of 56 images from the CTSpine1K dataset, respectively. Digitally reconstructed radiographs served as input for training and evaluation of the 2-dimensional (2D)-to-3-dimentional (3D) generative adversarial model. Model performance was assessed using peak signal to noise ratio (PSNR), structural similarity index (SSIM), and cosine similarity (CS). RESULTS: At external validation, the developed model achieved a PSNR of 21.139 ± 1.018 dB (mean ± standard deviation). The SSIM and CS amounted to 0.947 ± 0.010 and 0.671 ± 0.691, respectively. CONCLUSION: Generating an artificial 3D output from 2D imaging is challenging, especially for spinal imaging, where x-rays are known to deliver insufficient information frequently. Although the synthetic CT scans derived from our model do not perfectly match their ground truth CT, our proof-of-concept study warrants further exploration of the potential of this technology.

Safety and Effectiveness of an Enhanced Recovery Protocol in Patients Undergoing Burr Hole Evacuation for Chronic Subdural Hematoma.

BACKGROUND AND OBJECTIVES: Enhanced recovery programs may be especially useful in patients with chronic subdural hematoma or hygroma (cSDH), who frequently exhibit frailty and multimorbidity. We aim to evaluate the real-world safety and effectiveness of an enhanced recovery protocol in this population. METHODS: From a prospective registry, burr hole evacuations for cSDH carried out under the protocol (including early thromboprophylaxis, no flat bed rest, early mobilization without drain clamping, and early resumption of antithrombotic medication) were extracted, along with those procedures carried out within the past year before protocol change. Propensity score-based matching was carried out. A range of clinical and imaging outcomes were analyzed, including modified Rankin Scale as effectiveness and Clavien-Dindo adverse event grading as safety primary end points. RESULTS: Per group, 91 procedures were analyzed. At discharge, there was no significant difference in the modified Rankin Scale among the standard and enhanced recovery groups (1 [1; 2] vs 1 [1; 3], P = .552), or in Clavien-Dindo adverse event grading classifications of adverse events (P = .282) or occurrence of any adverse events (15.4% vs 20.9%, P = .442). There were no significant differences in time to drain removal (2.00 [2.00; 2.00] vs 2.00 [1.25; 2.00] days, P = .058), time from procedure to discharge (4.0 [3.0; 6.0] vs 4.0 [3.0; 6.0] days, P = .201), or total hospital length of stay (6.0 [5.0; 9.0] vs 5.0 [4.0; 8.0] days, P = .113). All-cause mortality was similar in both groups (8.8% vs 4.4%, P = .289), as was discharge disposition (P = .192). Other clinical and imaging outcomes were similar too (all P > .05). CONCLUSION: In a matched cohort study comparing perioperative standard of care with a novel enhanced recovery protocol focusing on evidence-based drainage, mobilization, and thromboprophylaxis regimens as well as changes to the standardized reuptake of oral anticoagulants and antiaggregants, no differences in safety or effectiveness were observed after burr hole evacuation of cSDH.