-
Machine-learning algorithms improved smoking identification in health records
Study comparing machine learning and rule-based algorithms for identifying smokers in administrative health data found ML models doubled sensitivity for detecting current smokers.
-
Vision–language model improved pediatric dental disease classification
Deep learning vision-language model for diagnosing pediatric dental diseases in panoramic radiographs, combining visual and textual information with 90% accuracy for caries and periapical.
-
Third molar PDL visibility helped classify age at 21
Study assesses periodontal ligament visibility in third molars for forensic age estimation at the 21-year legal threshold in Turkish population using ROC analysis.
-
Echocardiography may help identify high-risk pregnancy-induced hypertension patients
Echocardiographic parameters independently predict adverse pregnancy outcomes in women aged 35+ with pregnancy-induced hypertension, with combined assessment achieving excellent discrimination.
-
Value of histogram analysis of CT values in differential diagnosis of odontogenic keratocyst and ameloblastoma of the jaw
Histogram analysis of multislice CT attenuation values differentiates odontogenic keratocyst from ameloblastoma, with the 5th percentile parameter demonstrating optimal diagnostic performance.
