The Feasibility of a Machine Learning Approach in Predicting Successful Ventilator Mode Shifting for Adult Patients in the Medical Intensive Care Unit.

Background and Objectives: Traditional assessment of the readiness for the weaning from the mechanical ventilator (MV) needs respiratory parameters in a spontaneous breath. Exempted from the MV disconnecting and manual measurements of weaning parameters, a prediction model based on parameters from MV and electronic medical records (EMRs) may help the assessment before spontaneous breath trials. The study aimed to develop prediction models using machine learning techniques with parameters from the ventilator and EMRs for predicting successful ventilator mode shifting in the medical intensive care unit. Materials and Methods: A retrospective analysis of 1483 adult patients with mechanical ventilators for acute respiratory failure in three medical intensive care units between April 2015 and October 2017 was conducted by machine learning techniques to establish the predicting models. The input candidate parameters included ventilator setting and measurements, patients' demographics, arterial blood gas, laboratory results, and vital signs. Several classification algorithms were evaluated to fit the models, including Lasso Regression, Ridge Regression, Elastic Net, Random Forest, Extreme Gradient Boosting (XGBoost), Support Vector Machine, and Artificial Neural Network according to the area under the Receiver Operating Characteristic curves (AUROC). Results: Two models were built to predict the success shifting from full to partial support ventilation (WPMV model) or from partial support to the T-piece trial (sSBT model). In total, 3 MV and 13 nonpulmonary features were selected for the WPMV model with the XGBoost algorithm. The sSBT model was built with 8 MV and 4 nonpulmonary features with the Random Forest algorithm. The AUROC of the WPMV model and sSBT model were 0.76 and 0.79, respectively. Conclusions: The weaning predictions using machine learning and parameters from MV and EMRs have acceptable performance. Without manual measurements, a decision-making system would be feasible for the continuous prediction of mode shifting when the novel models process real-time data from MV and EMRs.

Development of a Three-Dimensional (3D) Printed Biodegradable Cage to Convert Morselized Corticocancellous Bone Chips into a Structured Cortical Bone Graft.

This study aimed to develop a new biodegradable polymeric cage to convert corticocancellous bone chips into a structured strut graft for treating segmental bone defects. A total of 24 adult New Zealand white rabbits underwent a left femoral segmental bone defect creation. Twelve rabbits in group A underwent three-dimensional (3D) printed cage insertion, corticocancellous chips implantation, and Kirschner-wire (K-wire) fixation, while the other 12 rabbits in group B received bone chips implantation and K-wire fixation only. All rabbits received a one-week activity assessment and the initial image study at postoperative 1 week. The final image study was repeated at postoperative 12 or 24 weeks before the rabbit scarification procedure on schedule. After the animals were sacrificed, both femurs of all the rabbits were prepared for leg length ratios and 3-point bending tests. The rabbits in group A showed an increase of activities during the first week postoperatively and decreased anterior cortical disruptions in the postoperative image assessments. Additionally, higher leg length ratios and 3-point bending strengths demonstrated improved final bony ingrowths within the bone defects for rabbits in group A. In conclusion, through this bone graft converting technique, orthopedic surgeons can treat segmental bone defects by using bone chips but with imitate characters of structured cortical bone graft.

Targeted concurrent and sequential delivery of chemotherapeutic and antiangiogenic agents to the brain by using drug-loaded nanofibrous membranes.

Glioblastoma is the most frequent and devastating primary brain tumor. Surgery followed by radiotherapy with concomitant and adjuvant chemotherapy is the standard of care for patients with glioblastoma. Chemotherapy is ineffective, because of the low therapeutic levels of pharmaceuticals in tumor tissues and the well-known tumor-cell resistance to chemotherapy. Therefore, we developed bilayered poly(d,l)-lactide-co-glycolide nanofibrous membranes that enabled the sequential and sustained release of chemotherapeutic and antiangiogenic agents by employing an electrospinning technique. The release characteristics of embedded drugs were determined by employing an in vitro elution technique and high-performance liquid chromatography. The experimental results showed that the fabricated nanofibers showed a sequential drug-eluting behavior, with the release of high drug levels of chemotherapeutic carmustine, irinotecan, and cisplatin from day 3, followed by the release of high concentrations of the antiangiogenic combretastatin from day 21. Biodegradable multidrug-eluting nanofibrous membranes were then dispersed into the cerebral cavity of rats by craniectomy, and the in vivo release characteristics of the pharmaceuticals from the membranes were investigated. The results suggested that the nanofibrous membranes released high concentrations of pharmaceuticals for more than 8 weeks in the cerebral parenchyma of rats. The result of histological analysis demonstrated developmental atrophy of brains with no inflammation. Biodegradable nanofibrous membranes can be manufactured for long-term sequential transport of different chemotherapeutic and anti-angiogenic agents in the brain, which can potentially improve the treatment of glioblastoma multiforme and prevent toxic effects due to systemic administration.

Combination of a biodegradable three-dimensional (3D) - printed cage for mechanical support and nanofibrous membranes for sustainable release of antimicrobial agents for treating the femoral metaphyseal comminuted fracture.

The aim of this study was to develop a biodegradable three-dimensional-printed polylactide (PLA) cage for promoting bony fixation and an antibiotics-embedded poly(d,l)-lactide-co-glycolide (PLGA) nanofibrous membrane for infectious prophylaxis during treating the comminuted metaphyseal fracture in a rabbit femoral model. The in vitro studies included measuring the mechanical properties of the 3D printed cage and determining release activities of vancomycin and ceftazidime from the nanofibers. The in vivo study included comparisons of rabbits of the femoral metaphyseal comminuted fracture treated with or without the combined biodegradable polymers. The results showed that vancomycin and ceftazidime were sustainably detected above the effective levels in the local tissue fluid around the fracture site for 3 weeks. The animal studies showed that rabbits with the 3D cage implantation possessed better cortical integrity, leg length ratio, and maximal bending strengths. The study results indicate that these combined polymers may promote fracture fixation during treating the rabbit femoral metaphyseal comminuted fracture.

Concurrent Chemotherapy of Malignant Glioma in Rats by Using Multidrug-Loaded Biodegradable Nanofibrous Membranes.

Glioblastoma multiforme has a poor prognosis and is highly chemoresistant. In this study, we implanted biodegradable 1,3-bis[2-chloroethyl]-1-nitroso-urea-, irinotecan-, and cisplatin-eluting poly[(d,l)-lactide-co-glycolide] (BIC/PLGA) and virgin nanofibrous membranes on the brain surface of C6 glioma-bearing rats in concurrent and virgin groups, respectively. The concentrations of all applied drugs were significantly higher in the brain than in the blood for more than 8 weeks in all studied rats. Tumor growth was more rapid in the vehicle-treated group, and tumor volumes were significantly higher in the vehicle-treated group. Moreover, the average survival time was significantly shorter in the vehicle-treated group (P = 0.026), and the BIC/PLGA nanofibrous membranes significantly reduced the risk of mortality (P < 0.001). Furthermore, the results suggested that the BIC/PLGA nanofibers reduced the malignancy of C6 glioma. The experimental findings indicate that the multianticancer drug (i.e., BIC)-eluting PLGA nanofibers are favorable candidates for treating malignant glioma.

Concurrent delivery of carmustine, irinotecan, and cisplatin to the cerebral cavity using biodegradable nanofibers: In vitro and in vivo studies.

Glioblastoma multiforme (GBM) is the most common primary malignant brain tumor, and the prognosis of patients afflicted with GBM has been dismal, exhibiting progressive neurologic impairment and imminent death. Even with the most active regimens currently available, chemotherapy achieves only modest improvement in the overall survival. New chemotherapeutic agents and novel approaches to therapy are required for improving clinical outcomes. In this study, we used an electrospinning technique and developed biodegradable poly[(d,l)-lactide-co-glycolide] nanofibrous membranes that facilitated a sustained release of carmustine (or bis-chloroethylnitrosourea, BCNU), irinotecan, and cisplatin. An elution method and a high-performance liquid chromatography assay were employed to characterize the in vitro and in vivo release behaviors of pharmaceuticals from the nanofibrous membranes. The experimental results showed that the biodegradable, nanofibrous membranes released high concentrations of BCNU, irinotecan, and cisplatin for more than 8 weeks in the cerebral cavity of rats. A histological examination revealed progressive atrophy of the brain tissues without inflammatory reactions. Biodegradable drug-eluting nanofibrous membranes may facilitate sustained delivery of various and concurrent chemotherapeutic agents in the cerebral cavity, enhancing the therapeutic efficacy of GBM treatment and preventing toxic effects resulting from the systemic administration of chemotherapeutic agents.