Regional Variation in Transient Ischemic Attack and Minor Stroke in Alberta Emergency Departments.

Background and Purpose- Multiple studies have shown the 90-day risk of stroke following an emergency department (ED) diagnosis of transient ischemic attack (TIA) or minor stroke is significant, with the greatest risk of recurrence being within the first 24 to 48 hours following initial symptom onset. This study explored regional differences in ED disposition, neuroimaging, and subsequent 90-day stroke risk of patients diagnosed with TIA or minor stroke in Alberta. Methods- We used administrative databases to identify ED visits, neuroimaging, and 90-day return visits for TIA or minor stroke in Alberta from April 2011 to March 2016 among adults ≥20 years of age and stratified them based on regions of presentation (Edmonton, Calgary, or nonmajor urban). Results- During the 5-year study period, 22 421 patients had index ED visits for TIA or minor stroke. All 3 regions had a similar number of ED visits for TIA/minor stroke; however, on index ED visit, Calgary had a higher proportion of computed tomographic angiography imaging (48.8%; P<0.0001) compared with Edmonton (6.7%) and nonmajor urban region (5.7%) and higher proportion of discharged patients (83%; P<0.0001) compared with Edmonton (77.7%) and nonmajor urban region (73.5%). The risk of admission for stroke within 90 days of discharge after index ED visit for TIA/minor stroke in Calgary (3.4%) was lower than Edmonton (4.5%) and the nonmajor urban region (4.6%; P=0.002). Conclusions- This study demonstrates regional variation in computed tomographic angiography for neurovascular imaging of patients presenting to the ED for TIA/minor stroke and a possible association with frequency of index visit admission and 90-day readmission for the same problem.

Predicting breast cancer-specific survival in metaplastic breast cancer patients using machine learning algorithms.

Metaplastic breast cancer (MpBC) is a rare and aggressive subtype of breast cancer, with data emerging on prognostic factors and survival prediction. This study aimed to develop machine learning models to predict breast cancer-specific survival (BCSS) in MpBC patients, utilizing a dataset of 160 patients with clinical, pathological, and biological variables. An in-depth variable selection process was carried out using gain ratio and correlation-based methods, resulting in 10 variables for model estimation. Five models (decision tree with bagging; logistic regression; multilayer perceptron; naïve Bayes; and, random forest algorithms) were evaluated using 10-fold cross-validation. Despite the constraints posed by the absence of therapeutic information, the random forest model exhibited the highest performance in predicting BCSS, with an ROC area of 0.808. This study emphasizes the potential of machine learning algorithms in predicting prognosis for complex and heterogeneous cancer subtypes using clinical datasets, and their potential to contribute to patient management. Further research that incorporates additional variables, such as treatment response, and more advanced machine learning techniques will likely enhance the predictive power of MpBC prognostic models.