Dataset Information

Predicting rotator cuff tears using data mining and Bayesian likelihood ratios.

ABSTRACT:

Objectives

Rotator cuff tear is a common cause of shoulder diseases. Correct diagnosis of rotator cuff tears can save patients from further invasive, costly and painful tests. This study used predictive data mining and Bayesian theory to improve the accuracy of diagnosing rotator cuff tears by clinical examination alone.

Methods

In this retrospective study, 169 patients who had a preliminary diagnosis of rotator cuff tear on the basis of clinical evaluation followed by confirmatory MRI between 2007 and 2011 were identified. MRI was used as a reference standard to classify rotator cuff tears. The predictor variable was the clinical assessment results, which consisted of 16 attributes. This study employed 2 data mining methods (ANN and the decision tree) and a statistical method (logistic regression) to classify the rotator cuff diagnosis into "tear" and "no tear" groups. Likelihood ratio and Bayesian theory were applied to estimate the probability of rotator cuff tears based on the results of the prediction models.

Results

Our proposed data mining procedures outperformed the classic statistical method. The correction rate, sensitivity, specificity and area under the ROC curve of predicting a rotator cuff tear were statistical better in the ANN and decision tree models compared to logistic regression. Based on likelihood ratios derived from our prediction models, Fagan's nomogram could be constructed to assess the probability of a patient who has a rotator cuff tear using a pretest probability and a prediction result (tear or no tear).

Conclusions

Our predictive data mining models, combined with likelihood ratios and Bayesian theory, appear to be good tools to classify rotator cuff tears as well as determine the probability of the presence of the disease to enhance diagnostic decision making for rotator cuff tears.

SUBMITTER: Lu HY

PROVIDER: S-EPMC3986413 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting rotator cuff tears using data mining and Bayesian likelihood ratios.

Lu Hsueh-Yi HY Huang Chen-Yuan CY Su Chwen-Tzeng CT Lin Chen-Chiang CC

PloS one 20140414 4

<h4>Objectives</h4>Rotator cuff tear is a common cause of shoulder diseases. Correct diagnosis of rotator cuff tears can save patients from further invasive, costly and painful tests. This study used predictive data mining and Bayesian theory to improve the accuracy of diagnosing rotator cuff tears by clinical examination alone.<h4>Methods</h4>In this retrospective study, 169 patients who had a preliminary diagnosis of rotator cuff tear on the basis of clinical evaluation followed by confirmator ...[more]

PMID: 24733553

Similar Datasets

Project description:PurposeTo determine whether anterior cable tears could be identified at the time of arthroscopic rotator cuff repair and determine the characteristics of the anterior cable tears identified.MethodsFrom 2016 to 2017 all shoulder arthroscopies had data collected prospectively at the time of surgery, specifically including injury to the capsular and tendon zones of insertion on the greater tuberosity. Anterior cable position and degree of injury and medialization were recorded, as well as complete findings of the diagnostic arthroscopy. The inclusion criterion was primary shoulder arthroscopy. The exclusion criterion was any revision shoulder arthroscopy. All arthroscopic rotator cuff repairs (ARCR) were grouped together and all other nonarthroscopic rotator cuff repair surgeries (non-ARCR) were grouped together.ResultsIn total, 118 shoulder arthroscopies had data collected prospectively at the time of surgery: 90 primary shoulder arthroscopies met the inclusion criteria; 28 were excluded because they were revision surgeries. There were 42 patients in the ARCR group (Group 1). Six of these were partial tears, and 36 were full-thickness tears. There were 48 patients in the non-ARCR group (Group 2). The non-ARCR Group 2 served as an anatomic baseline for ARCR Group 1. In all 90 shoulders, the rotator cable and anterior cable were identified. Group 1 (ARCR) incidence of anterior cable tears with abnormal position was 71.4% compared to 2.1% in group 2 (non-ARCR) (P < .001) Group 1 (ARCR) incidence of anterior cable tears with normal anterior cable position (n = 12) was compared to abnormal anterior cable position (n = 30). Injury to the anterior footprint capsular and tendon zones were compared. Normal anterior cable position correlated with no or low-grade injury to anterior footprint capsular zone. (Nimura zone C1). Abnormal anterior cable displacement graded as moderate (n = 20) and severe (n = 10) were compared for injury to the anterior footprint. Moderate displacement correlated with complete or high grade injury to C1 in 85% and complete injury to R1 in 45% (P < .001 and .049). In severe displacement complete C1 injury was 100%, and complete R1 injury was 100% (P < .001 and .001).ConclusionsAnterior cable tears are universally identified in ARCR. Three patterns of medial displacement severity correlated with injury to a crucial insertion zone (C1) at the anterior footprint. The degree of anterior cable disruption at the anterior footprint and displacement was commonly disproportionately greater than the injury to the supraspinatus.Level of evidenceLevel III, diagnostic study.

Dataset Information

Predicting rotator cuff tears using data mining and Bayesian likelihood ratios.

Objectives

Methods

Results

Conclusions

Publications

Predicting rotator cuff tears using data mining and Bayesian likelihood ratios.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets