Machine learning prediction models for in-hospital postoperative functional outcome after moderate-to-severe traumatic brain injury.

AIM: This study aims to utilize machine learning (ML) and logistic regression (LR) models to predict surgical outcomes among patients with traumatic brain injury (TBI) based on admission examination, assisting in making optimal surgical treatment decision for these patients. METHOD: We conducted a retrospective review of patients hospitalized in our department for moderate-to-severe TBI. Patients admitted between October 2011 and October 2022 were assigned to the training set, while patients admitted between November 2022 and May 2023 were designated as the external validation set. Five ML algorithms and LR model were employed to predict the postoperative Glasgow Outcome Scale (GOS) status at discharge using clinical and routine blood data collected upon admission. The Shapley (SHAP) plot was utilized for interpreting the models. RESULTS: A total of 416 patients were included in this study, and they were divided into the training set (n = 396) and the external validation set (n = 47). The ML models, using both clinical and routine blood data, were able to predict postoperative GOS outcomes with area under the curve (AUC) values ranging from 0.860 to 0.900 during the internal cross-validation and from 0.801 to 0.890 during the external validation. In contrast, the LR model had the lowest AUC values during the internal and external validation (0.844 and 0.567, respectively). When blood data was not available, the ML models achieved AUCs of 0.849 to 0.870 during the internal cross-validation and 0.714 to 0.861 during the external validation. Similarly, the LR model had the lowest AUC values (0.821 and 0.638, respectively). Through repeated cross-validation analysis, we found that routine blood data had a significant association with higher mean AUC values in all ML and LR models. The SHAP plot was used to visualize the contributions of all predictors and highlighted the significance of blood data in the lightGBM model. CONCLUSION: The study concluded that ML models could provide rapid and accurate predictions for postoperative GOS outcomes at discharge following moderate-to-severe TBI. The study also highlighted the crucial role of routine blood tests in improving such predictions, and may contribute to the optimization of surgical treatment decision-making for patients with TBI.

Machine learning models for predicting in-hospital outcomes after non-surgical treatment among patients with moderate-to-severe traumatic brain injury.

AIM: This study aims to develop prediction models for in-hospital outcomes after non-surgical treatment among patients with moderate-to-severe traumatic brain injury (TBI). METHOD: We conducted a retrospective review of patients hospitalized for moderate-to-severe TBI in our department from 2011 to 2020. Five machine learning (ML) algorithms and the conventional logistic regression (LR) model were employed to predict in-hospital mortality and the Glasgow Outcome Scale (GOS) functional outcomes. These models utilized clinical and routine blood data collected upon admission. RESULTS: This study included a total of 196 patients who received only non-surgical treatment after moderate-to-severe TBI. When predicting mortality, ML models achieved area under the curve (AUC) values of 0.921 to 0.994 using clinical and routine blood data, and 0.877 to 0.982 using only clinical data. In comparison, LR models yielded AUCs of 0.762 and 0.730 respectively. When predicting the GOS outcome, ML models achieved AUCs of 0.870 to 0.915 using clinical and routine blood data, and 0.858 to 0.927 using only clinical data. In comparison, the LR model yielded AUCs of 0.798 and 0.787 respectively. Repeated internal validation showed that the contributions of routine blood data for prediction models may depend on different prediction algorithms and different outcome measurements. CONCLUSION: The study reported ML-based prediction models that provided rapid and accurate predictions on short-term outcomes after non-surgical treatment among patients with moderate-to-severe TBI. The study also highlighted the superiority of ML models over conventional LR models and proposed the complex contributions of routine blood data in such predictions.

Correlation between intraosseous thermal change and drilling impulse data during osteotomy within autonomous dental implant robotic system: An in vitro study.

OBJECTIVES: This study aims at examining the correlation of intraosseous temperature change with drilling impulse data during osteotomy and establishing real-time temperature prediction models. MATERIALS AND METHODS: A combination of in vitro bovine rib model and Autonomous Dental Implant Robotic System (ADIR) was set up, in which intraosseous temperature and drilling impulse data were measured using an infrared camera and a six-axis force/torque sensor respectively. A total of 800 drills with different parameters (e.g., drill diameter, drill wear, drilling speed, and thickness of cortical bone) were experimented, along with an independent test set of 200 drills. Pearson correlation analysis was done for linear relationship. Four machining learning (ML) algorithms (e.g., support vector regression [SVR], ridge regression [RR], extreme gradient boosting [XGboost], and artificial neural network [ANN]) were run for building prediction models. RESULTS: By incorporating different parameters, it was found that lower drilling speed, smaller drill diameter, more severe wear, and thicker cortical bone were associated with higher intraosseous temperature changes and longer time exposure and were accompanied with alterations in drilling impulse data. Pearson correlation analysis further identified highly linear correlation between drilling impulse data and thermal changes. Finally, four ML prediction models were established, among which XGboost model showed the best performance with the minimum error measurements in test set. CONCLUSION: The proof-of-concept study highlighted close correlation of drilling impulse data with intraosseous temperature change during osteotomy. The ML prediction models may inspire future improvement on prevention of thermal bone injury and intelligent design of robot-assisted implant surgery.

Genome-wide DNA methylation analysis identifies potent CpG signature for temzolomide response in non-G-CIMP glioblastomas with unmethylated MGMT promoter: MGMT-dependent roles of GPR81.

PURPOSES: To identify potent DNA methylation candidates that could predict response to temozolomide (TMZ) in glioblastomas (GBMs) that do not have glioma-CpGs island methylator phenotype (G-CIMP) but have an unmethylated promoter of O-6-methylguanine-DNA methyltransferase (unMGMT). METHODS: The discovery-validation approach was planned incorporating a series of G-CIMP-/unMGMT GBM cohorts with DNA methylation microarray data and clinical information, to construct multi-CpG prediction models. Different bioinformatic and experimental analyses were performed for biological exploration. RESULTS: By analyzing discovery sets with radiotherapy (RT) plus TMZ versus RT alone, we identified a panel of 64 TMZ efficacy-related CpGs, from which a 10-CpG risk signature was further constructed. Both the 64-CpG panel and the 10-CpG risk signature were validated showing significant correlations with overall survival of G-CIMP-/unMGMT GBMs when treated with RT/TMZ, rather than RT alone. The 10-CpG risk signature was further observed for aiding TMZ choice by distinguishing differential outcomes to RT/TMZ versus RT within each risk subgroup. Functional studies on GPR81, the gene harboring one of the 10 CpGs, indicated its distinct impacts on TMZ resistance in GBM cells, which may be dependent on the status of MGMT expression. CONCLUSIONS: The 64 TMZ efficacy-related CpGs and in particular the 10-CpG risk signature may serve as promising predictive biomarker candidates for guiding optimal usage of TMZ in G-CIMP-/unMGMT GBMs.

Machine learning algorithms for improved prediction of in-hospital outcomes after moderate-to-severe traumatic brain injury: a Chinese retrospective cohort study.

AIM: Controversy remains high over the superiority of advanced machine learning (ML) algorithms to conventional logistic regression (LR) in the prediction of prognosis after traumatic brain injury (TBI). This study aimed to compare the performance of ML and LR models in predicting in-hospital prognosis after TBI. METHOD: In a single-center retrospective cohort of adult patients hospitalized for moderate-to-severe TBI (Glasgow coma score ≤12) in our hospital from 2011 to 2020, LR and three ML algorithms (XGboost, lightGBM, and FT-transformer) were run to build prediction models for in-hospital mortality and the Glasgow Outcome Scale (GOS) functional outcomes using either all 19 clinical and laboratory features or the 10 non-laboratory ones collected at admission to the neurological intensive care unit. The Shapley (SHAP) value was used for model interpretation. RESULT: In total, 482 patients had an in-hospital mortality rate of 11.0%. A total of 23.0% of the patients had good functional scores (GOS ≥ 4) at discharge. All ML models performed better than the LR model in predicting in-hospital prognosis after TBI, among which the lightGBM model showed the best performance: When predicting mortality, the lightGBM model yielded an area under the curve (AUC) of 0.953 using all 19 features (the LR model: 0.813) and an AUC of 0.935 using 10 non-laboratory features (the LR model: 0.803); when predicting GOS functional outcomes, it yielded an AUC of 0.913 using all 19 features (the LR model: 0.832) and an AUC of 0.889 using non-laboratory data (the LR model: 0.818). The SHAP method identified key contributors to explain the lightGBM models. Finally, the integration of the lightGBM models with different prediction purposes was found to provide refined prognostic information, particularly for patients who survived moderate-to-severe TBI. CONCLUSION: The study supported the superiority of ML to LR in predicting prognosis after moderate-to-severe TBI and highlighted its potential use for clinical application.

Novel predictive epigenetic signature for temozolomide in non-G-CIMP glioblastomas.

OBJECTIVE: To identify novel epigenetic signatures that could provide predictive information that is complementary to promoter methylation status of the O-6-methylguanine-DNA methyltransferase (MGMT) gene for predicting temozolomide (TMZ) response, among glioblastomas (GBMs) without glioma-CpGs island methylator phenotype (G-CIMP) METHODS: Different cohorts of primary non-G-CIMP GBMs with genome-wide DNA methylation microarray data were included for discovery and validation of a multimarker signature, combined using a RISK score model. Different statistical analyses and functional experiments were performed for clinical and biological validation. RESULTS: By employing discovery cohorts with radiotherapy (RT) and TMZ versus RT alone and a strict multistep selection strategy, we identified seven CpGs, each of which was significantly correlated with overall survival (OS) of non-G-CIMP GBMs with RT/TMZ, independent of age, MGMT promoter methylation status, and other identified CpGs. A RISK score signature of the 7 CpGs was developed and validated to distinguish non-G-CIMP GBMs with differential survival outcomes to RT/TMZ, but not to RT alone. The interaction analyses also showed differential outcomes to RT/TMZ versus RT alone within the RISK score-based subgroups. The signature could also improve the risk classification by age and MGMT promoter methylation status. Functional experiments showed that HSBP2 appeared to be epigenetically regulated by one identified CpG and was associated with TMZ resistance, but it was not associated with cell proliferation or apoptosis in GBM cell lines. The predictive value of the single CpG methylation of HSBP2 by pyrosequencing was observed in a local cohort of isocitrate dehydrogenase 1 (IDH1) R132H wild-type GBMs. CONCLUSIONS: This novel epigenetic signature might be a promising predictive (but not a general prognostic) biomarker and be helpful for refining the MGMT-based guiding approach to TMZ usage in non-G-CIMP GBMs.

A five-CpG signature of microRNA methylation in non-G-CIMP glioblastoma.

AIMS: DNA methylation has been found to regulate microRNAs (miRNAs) expression, but the prognostic value of miRNA-related DNA methylation aberration remained largely elusive in cancers including glioblastomas (GBMs). This study aimed to investigate the clinical and biological feature of miRNA methylation in GBMs of non-glioma-CpG island methylator phenotype (non-G-CIMP). METHODS: Prognostic miRNA methylation loci were analyzed, with TCGA and Rennes cohort as training sets, and independent datasets of GBMs and low-grade gliomas (LGGs) were obtained as validation sets. Different statistical and bioinformatic analysis and experimental validations were performed to clinically and biologically characterize the signature. RESULTS: We identified and validated a risk score based on methylation status of five miRNA-associated CpGs which could predict survival of GBM patients in a series of training and validation sets. This signature was independent of age and O-6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status. The risk subgroup was associated with angiogenesis and accordingly differential responses to bevacizumab-contained therapy. MiRNA target analysis and in vitro experiments further confirmed the accuracy of this signature. CONCLUSION: The five-CpG signature of miRNA methylation was biologically relevant and was of potential prognostic and predictive value for GBMs. It might be of help for improving individualized treatment.

A novel prognostic six-CpG signature in glioblastomas.

AIMS: We aimed to identify a clinically useful biomarker using DNA methylation-based information to optimize individual treatment of patients with glioblastoma (GBM). METHODS: A six-CpG panel was identified by incorporating genome-wide DNA methylation data and clinical information of three distinct discovery sets and was combined using a risk-score model. Different validation sets of GBMs and lower-grade gliomas and different statistical methods were implemented for prognostic evaluation. An integrative analysis of multidimensional TCGA data was performed to molecularly characterize different risk tumors. RESULTS: The six-CpG risk-score signature robustly predicted overall survival (OS) in all discovery and validation cohorts and in a treatment-independent manner. It also predicted progression-free survival (PFS) in available patients. The multimarker epigenetic signature was demonstrated as an independent prognosticator and had better performance than known molecular indicators such as glioma-CpG island methylator phenotype (G-CIMP) and proneural subtype. The defined risk subgroups were molecularly distinct; high-risk tumors were biologically more aggressive with concordant activation of proangiogenic signaling at multimolecular levels. Accordingly, we observed better OS benefits of bevacizumab-contained therapy to high-risk patients in independent sets, supporting its implication in guiding usage of antiangiogenic therapy. Finally, the six-CpG signature refined the risk classification based on G-CIMP and MGMT methylation status. CONCLUSIONS: The novel six-CpG signature is a robust and independent prognostic indicator for GBMs and is of promising value to improve personalized management.

Dec1 expression predicts prognosis and the response to temozolomide chemotherapy in patients with glioma.

Differentiated embryo chondrocyte expressed gene 1 (Dec1), a crucial cell differentiation mediator and apoptosis inhibitor, is abundantly expressed in various types of human cancer and is associated with malignant tumor progression. As poor differentiation and low apoptosis are closely associated with poor survival rates and a poor response to radio/chemotherapy in patients with cancer, the prognostic value of Dec1 expression was examined in the present study and its correlation with response to temozolomide (TMZ) chemotherapy was analyzed in patients with glioma. Dec1 expression was analyzed by immunohistochemistry in 157 samples of newly diagnosed glioma and 63 recurrent glioblastoma cases that relapsed during TMZ chemotherapy. Correlations with clinical variables, prognosis and the response to TMZ chemotherapy were analyzed in the newly diagnosed gliomas. Dec1 expression was also compared with the apoptosis index determined by TdT­mediated dUTP nick ending­labeling assay in recurrent glioblastomas. The antiglioma effect of TMZ in nude mice xenografts with Dec1 expression was examined in vivo. High expression of Dec1, which was significantly associated with high pathological tumor grade and poor response to TMZ chemotherapy, was demonstrated to be an unfavorable independent prognostic factor and predicted poor survival in patients with newly diagnosed glioma. In patients with recurrent glioblastoma, there was a negative correlation between Dec1 expression and the apoptotic index. In nude mice treated with TMZ, Dec1 overexpression potentiated proliferation, but attenuated TMZ­induced apoptosis. In conclusion, Dec1 is a prognostic factor for the clinical outcome and a predictive factor for the response to TMZ chemotherapy in patients with glioma.

New way to achieve frontal facial photographs of the head in its natural position.

Frontal photographs of the head in its natural position are not as easy to achieve as lateral ones. We describe a new way to obtain standard 2-dimensional images of its natural position in full-face frontal view using a customised photographic system, which may provide supplementary information for traditional lateral facial imaging, and be helpful for standard assessment of 3-dimensional facial images.

The predictive but not prognostic value of MGMT promoter methylation status in elderly glioblastoma patients: a meta-analysis.

BACKGROUND: The clinical implication of O6-methylguanine-DNA methyltransferase (MGMT) promoter status is ill-defined in elderly glioblastoma patients. Here we report a meta-analysis to seek valid evidence for its clinical relevance in this subpopulation. METHODS: Literature were searched and reviewed in a systematic manner using the PubMed, EMBASE and Cochrane databases. Studies investigating the association between MGMT promoter status and survival data of elderly patients (≥65 years) were eligible for inclusion. RESULTS: Totally 16 studies were identified, with 13 studies included in the final analyses. The aggregate proportion of MGMT promoter methylation in elderly patients was 47% (95% confidence interval [CI]: 42-52%), which was similar to the value for younger patients. The analyses showed differential effects of MGMT status on overall survival (OS) of elderly patients according to assigned treatments: methylated vs. unmethylated: (1) temozolomide (TMZ)-containing therapies: hazard ratio [HR] 0.49, 95% CI 0.41-0.58; (2) TMZ-free therapies: HR 0.97, 95% CI 0.77-1.21. More importantly, a useful predictive value was observed by an interaction analysis: TMZ-containing therapies vs. RT alone: (1) methylated tumors: HR 0.48, 95% CI 0.36-0.65; (2) unmethylated tumors: HR 1.14; 95% CI 0.90-1.44. CONCLUSION: The meta-analysis reports an age-independent presence of MGMT promoter methylation. More importantly, the study encouraged routine testing of MGMT promoter status especially in elderly glioblastoma patients by indicating a direct linkage between biomarker test and individual treatment decision. Future studies are needed to justify the mandatory testing in younger patients.

A meta-analysis of temozolomide versus radiotherapy in elderly glioblastoma patients.

Temozolomide (TMZ) alone has been proposed as a promising alternative to radiotherapy (RT) in elderly glioblastoma (GBM) patients. We report a meta-analysis to systematically evaluate TMZ monotherapy in older GBM patients. A systematic literature search was performed using PubMed, EMBASE and the Cochrane database. Studies comparing TMZ versus RT in elderly patients (≥ 65 years) with newly diagnosed GBM were eligible for inclusion. Two randomized clinical trials (RCTs) and three comparative studies were included in the analyses, which revealed an overall survival (OS) advantage for TMZ compared with RT (HR [hazard ratio] 0.86, 95 % CI [confidence interval] 0.74-1.00). However, a sensitivity analysis of 2 RCTs only supported its non-inferiority (HR 0.91, 95 % CI 0.66-1.27). Most elderly patients tolerated TMZ despite an increased risk of grade 3-4 (G3-4) toxicities, especially hematological toxicities. The quality of life was similar between the groups. In the MGMT analysis, methylated tumors were associated with a longer OS than unmethylated tumors among elderly patients receiving TMZ monotherapy (HR 0.50, 95 % CI 0.35-0.70). Moreover, in patients with methylated tumors, TMZ was more beneficial than RT alone in improving OS (TMZ vs. RT: HR 0.66, 95 % CI 0.47-0.93) whereas the opposite was true for those with unmethylated tumors (HR 1.32, 95 % CI 1.00-1.76). Although the meta-analysis demonstrated the non-inferiority to RT in improving OS, TMZ alone was not a straightforward solution for elderly GBM patients because of an increased risk of G3-4 toxicities, especially hematological toxicities. MGMT testing might be helpful for determining individualized treatment.

Radiotherapy plus concurrent or sequential temozolomide for glioblastoma in the elderly: a meta-analysis.

BACKGROUND: Many physicians are reluctant to treat elderly glioblastoma (GBM) patients as aggressively as younger patients, which is not evidence based due to the absence of validated data from primary studies. We conducted a meta-analysis to provide valid evidence for the use of the aggressive combination of radiotherapy (RT) and temozolomide (TMZ) in elderly GBM patients. METHODS: A systematic literature search was conducted using the PubMed, EMBASE and Cochrane databases. Studies comparing combined RT/TMZ with RT alone in elderly patients (≥65 years) with newly diagnosed GBM were eligible for inclusion. RESULTS: No eligible randomized trials were identified. Alternatively, a meta-analysis of nonrandomized studies (NRSs) was performed, with 16 studies eligible for overall survival (OS) analysis and nine for progression-free survival (PFS) analysis. Combined RT/TMZ was shown to reduce the risk of death and progression in elderly GBM patients compared with RT alone (OS hazard ratio [HR] 0.59, 95% confidence interval [CI] 0.48-0.72; PFS: HR 0.58, 95% CI 0.41-0.84). Evaluable patients were reported to tolerate combined treatment but certain toxicities, and especially hematological toxicities, were more frequently observed. Limited data on O6-methylguanine-DNA methyltransferase (MGMT) promoter status and quality of life were reported. CONCLUSION: The meta-analysis of NRSs provided level 2a evidence (Oxford Centre for Evidence-Based Medicine) that combined RT/TMZ conferred a clear survival benefit on a selection of elderly GBM patients who had a favorable prognosis (e.g., extensive resection, favorable KPS). Toxicities were more frequent but acceptable. Future randomized trials are warranted to justify a definitive conclusion.

The treatment of glioblastomas: a systematic update on clinical Phase III trials.

Glioblastomas (GBMs) are invariably associated with unavoidable tumor recurrence and overall poor prognosis. The present study is to summarize the results of clinical Phase III studies on GBMs over the past seven years. A systematic literature search was performed using major electronic databases and by screening meeting abstracts. Totally, 16 studies of patients with newly diagnosed GBMs, recurrent GBMs, and elderly patients with GBMs were selected for this review. Although the outcomes of the experimental therapies were not encouraging, these studies produced a considerable amount of potentially clinically relevant information. Such aspects as surgical outcomes, radiation schedules, temozolomide (TMZ) schedules, methylation status of the O6-methylguanine DNA methyltransferase (MGMT) gene, combination of therapies, novel drug delivery methods and use of targeted agents have come to light and are being addressed here. In addition, we discuss the existing controversies of (1) surgical studies, (2) evaluations of recurrence, (3) salvage treatment bias, and (4) studies on elderly patients.