ABSTRACT
BACKGROUND: Schizophrenia is a severe psychiatric disease, and its prevalence is higher. However, diagnosis of early-stage schizophrenia is still considered a challenging task. AIM: To employ brain morphological features and machine learning method to differentiate male individuals with schizophrenia from healthy controls. METHODS: The least absolute shrinkage and selection operator and t tests were applied to select important features from structural magnetic resonance images as input features for classification. Four commonly used machine learning algorithms, the general linear model, random forest (RF), k-nearest neighbors, and support vector machine algorithms, were used to develop the classification models. The performance of the classification models was evaluated according to the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 8 important features with significant differences between groups were considered as input features for the establishment of classification models based on the four machine learning algorithms. Compared to other machine learning algorithms, RF yielded better performance in the discrimination of male schizophrenic individuals from healthy controls, with an AUC of 0.886. CONCLUSION: Our research suggests that brain morphological features can be used to improve the early diagnosis of schizophrenia in male patients.
ABSTRACT
<p><b>OBJECTIVE</b>To compare the diagnostic performance of prospective electrocardiogram (ECG)-triggered computed tomography coronary angiography (CTCA) versus retrospective ECG-gated CTCA.</p><p><b>METHODS</b>Patients with suspected coronary artery disease divided into two groups which underwent 64-slice CTCA with prospective ECG-triggered or retrospective ECG-gated scanning (n = 100 each, HR < or = 65 bpm). Multi-planar reconstruction (MPR), curved-planar reconstruction (CPR), maximum intensity projection (MIP) and volume rendering (VR) were made to demonstrate the coronary arteries. The image quality was defined as excellent, good and poor by motion and stair-step artifacts. Individual radiation exposure dose was estimated from the dose-length product.</p><p><b>RESULTS</b>The mean effective radiation dose of prospective ECG-triggered CTCA [(2.81 +/- 0.48) mSv] was significantly lower than that of retrospective ECG-gated CTCA [(10.16 +/- 1.09) mSv, P < 0.01]. Segments of diagnostic image quality (95.2%, 1165/1224) and non-diagnostic coronary segments (4.8%, 59/1224) in prospective ECG-triggered group were similar as those of retrospective ECG-gated group (94.1%, 1186/1261 and 5.9%, 75/1261, all P > 0.05).</p><p><b>CONCLUSION</b>Though the radiation exposure dose required is significantly lower, the diagnostic performance of prospective ECG-triggered 64 slice CTCA is comparable with that of retrospective ECG-gated 64 slice CTCA on patients with stable heart rates up to 65 bpm.</p>
