Use of Machine Learning Theory to Predict the Need for Femoral Nerve Block Following ACL Repair

Tighe, Patrick J.; Laduzenski, Sarah; Edwards, David A.; Ellis, Neal; Boezaart, André P.; Aygtug, Haldun

doi:10.1111/j.1526-4637.2011.01228.x

Cited by 27 publications

(21 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our results suggest that predictive analytics techniques, such as MLCs, may offer considerable assistance as clinical decision support tools, but must be compared and validated using multiple metrics of classifier performance. The results of this study complement previous work demonstrating the potential utility of MLC to forecast the need for a postoperative rescue femoral nerve block following anterior cruciate ligament repair [22].…”

Section: Discussionsupporting

confidence: 81%

Use of Machine-Learning Classifiers to Predict Requests for Preoperative Acute Pain Service Consultation

Tighe¹,

Lucas²,

Edwards³

et al. 2012

Pain Med

View full text Add to dashboard Cite

Objective The purpose of this project was to determine whether machine-learning classifiers could predict which patients would require a preoperative acute pain service consultation. Design Retrospective cohort. Setting University teaching hospital. Subjects The records of 9,860 surgical patients posted between January 1 and June 30, 2010 were reviewed. Outcome Measures Request for acute pain service consultation. A cohort of machine-learning classifiers was compared according to its ability or inability to classify surgical cases as requiring a request for a preoperative acute pain service consultation. Classifiers were then optimized utilizing ensemble techniques. Computational efficiency was measured with the central processing unit processing times required for model training. Classifiers were tested using the full feature set, as well as the reduced feature set that was optimized using a merit-based dimensional reduction strategy. Results Machine-learning classifiers correctly predicted preoperative requests for acute pain service consultations in 92.3% (95% confidence intervals [CI], 91.8–92.8) of all surgical cases. Bayesian methods yielded the highest area under the receiver operating curve (0.87, 95% CI 0.84–0.89) and lowest training times (0.0018 seconds, 95% CI, 0.0017–0.0019 for the NaiveBayesUpdateable algorithm). An ensemble of high-performing machine-learning classifiers did not yield a higher area under the receiver operating curve than its component classifiers. Dimensional reduction decreased the computational requirements for multiple classifiers, but did not adversely affect classification performance. Conclusions Using historical data, machine-learning classifiers can predict which surgical cases should prompt a preoperative request for an acute pain service consultation. Dimensional reduction improved computational efficiency and preserved predictive performance.

show abstract

Section: Discussionsupporting

confidence: 81%

Use of Machine-Learning Classifiers to Predict Requests for Preoperative Acute Pain Service Consultation

Tighe¹,

Lucas²,

Edwards³

et al. 2012

Pain Med

View full text Add to dashboard Cite

show abstract

“…This algorithm also differs from standard ‘if–then’ decision trees and classification and regression trees (CART). The ADTree method has several advantages compared with these other machine learning algorithms, including: (1) several comparative studies have shown higher accuracy and versatility for ADTree than other machine learning methods [18,19]; and (2) the ADTree model structure is less complex than other methods [16], which facilitates model interpretation and reduces the need for model optimization.…”

Section: Introductionmentioning

confidence: 99%

Prediction of axillary lymph node metastasis in primary breast cancer patients using a decision tree-based model

Takada

Sugimoto

Naito

et al. 2012

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

BackgroundThe aim of this study was to develop a new data-mining model to predict axillary lymph node (AxLN) metastasis in primary breast cancer. To achieve this, we used a decision tree-based prediction method—the alternating decision tree (ADTree).MethodsClinical datasets for primary breast cancer patients who underwent sentinel lymph node biopsy or AxLN dissection without prior treatment were collected from three institutes (institute A, n = 148; institute B, n = 143; institute C, n = 174) and were used for variable selection, model training and external validation, respectively. The models were evaluated using area under the receiver operating characteristics (ROC) curve analysis to discriminate node-positive patients from node-negative patients.ResultsThe ADTree model selected 15 of 24 clinicopathological variables in the variable selection dataset. The resulting area under the ROC curve values were 0.770 [95% confidence interval (CI), 0.689–0.850] for the model training dataset and 0.772 (95% CI: 0.689–0.856) for the validation dataset, demonstrating high accuracy and generalization ability of the model. The bootstrap value of the validation dataset was 0.768 (95% CI: 0.763–0.774).ConclusionsOur prediction model showed high accuracy for predicting nodal metastasis in patients with breast cancer using commonly recorded clinical variables. Therefore, our model might help oncologists in the decision-making process for primary breast cancer patients before starting treatment.

show abstract

“…SVMs have been successfully applied to many classification and function prediction tasks, in various fields including science, engineering, and social sciences . SVM‐based classification methods have also been used for surgical applications, while similar regression techniques have been used to a comparatively lesser extent in medical applications . Our analysis was performed using LibSVM (National Taiwan University, http://www.csie.ntu.edu.tw/∼cjlin/libsvm/).…”

Section: Methodsmentioning

confidence: 99%

Bone mineral density and donor age are not predictive of femoral ring allograft bone mechanical strength

2014

View full text Add to dashboard Cite

While metal or plastic interbody spinal fusion devices are manufactured to appropriate mechanical standards, mechanical properties of commercially prepared structural allograft bone remain relatively unassessed. Robust models predicting compressive load to failure of structural allograft bone based on easily measured variables would be useful. Three hundred twenty seven femoral rings from 34 cadaver femora were tested to failure in axial compression. Predictive variables included age, gender, bone mineral density (BMD), position along femoral shaft, maximum/minimum wall thickness, outer/inner diameter, and area. We used support vector regression and 10-fold cross-validation to develop robust nonlinear predictive models for load to failure. Model performance was measured by the root-mean-squared-deviation (RMSD) and correlation coefficients (r). A polynomial model using all variables had RMSD ¼ 7.92, r ¼ 0.84, indicating excellent performance. A model using all variables except BMD was essentially unchanged (RMSD ¼ 8.12, r ¼ 0.83). Eliminating both age and BMD produced a model with RMSD ¼ 8.41, r ¼ 0.82, again essentially unchanged. Compressive strength of structural allograft bone can be estimated using easily measured geometric parameters, without including BMD or age. As DEXA is costly and cumbersome, and setting upper age-limits for potential donors reduces the supply, our results may prove helpful to increase the quality and availability of structural allograft. Keywords: structural bone allograft; spinal fusion; biomechanics; bone mineral densityThe allograft bone industry is guided by practices intended to optimize safety and effectiveness of allograft bone in numerous clinical applications. Guidelines have been developed and used with respect to avoidance of transmission of neoplastic and infectious disease based on high level scientific data. 1,2 Adherence to these guidelines has resulted in safe and successful allograft implantation over many years of experience. 3 Despite this excellent track record, the mechanical properties of what is essentially a mechanical device 4 remain comparatively un-assessed. In fact, there are no guidelines prescribed with respect to material properties of structural allograft bone products within the Food and Drug Administration or the American Association of Tissue Banks. 2,5 A recent survey of allograft providers demonstrated that given this lack of guidelines, a variety of practices are employed with regard to screening of donors and allografts. 6 On the other hand, interbody spacers manufactured using materials such as carbon fiber, titanium, or polyetheretherketone are expected to have consistent biomechanical properties. 7-9 While the cited advantages of allograft bone over such implants include lower cost and equal or better clinical performance, 10-21 the mechanical performance of structural allograft bone intended for use as an interbody device should be similarly uniform and reliable.We have accumulated a database of load to failure measurements for fem...

show abstract

Use of Machine Learning Theory to Predict the Need for Femoral Nerve Block Following ACL Repair

Abstract: ML classifiers may offer improved predictive capabilities when analyzing medical data sets compared with traditional statistical methodologies in predicting severe postoperative pain requiring peripheral nerve block.

Cited by 27 publications

References 25 publications

Use of Machine-Learning Classifiers to Predict Requests for Preoperative Acute Pain Service Consultation

Use of Machine-Learning Classifiers to Predict Requests for Preoperative Acute Pain Service Consultation

Prediction of axillary lymph node metastasis in primary breast cancer patients using a decision tree-based model

Bone mineral density and donor age are not predictive of femoral ring allograft bone mechanical strength

Contact Info

Product

Resources

About