Clinical value of synthetic MRI and dynamic contrast-enhanced MRI in the differentiation of benign and malignant breast lesions / 中华放射学杂志

Objective:To explore the value of synthetic MRI and dynamic contrast-enhanced MRI (DCE-MRI) in differentiating benign and malignant breast lesions.Methods:Totally 116 patients (female, aged 22-80 years) with suspicious breast lesions found by clinical examination in General Hospital of Ningxia Medical University from September 2020 to March 2021 were retrospectively analyzed. All patients received synthetic MRI and DCE-MRI scans, and the quantitative parameters of synthetic MRI [T 1, T 2, proton density (PD) value before enhancement of breast lesions (T 1pre, T 2pre, PD pre) and T 1, T 2, PD value after enhancement of breast lesions (T 1Gd, T 2Gd, PD Gd)] were measured. The difference before and after enhancement (?T 1, ?T 2, ?PD) and the ratio before and after enhancement(T 1ratio, T 2ratio, PD ratio) were calculated. All lesions were confirmed by biopsy or surgery. Independent sample t test was used to compare the differences of parameters between benign and malignant lesions, and parameters with statistically significant differences were analyzed by multivariate logistic regression to screen out independent predictors for the diagnosis of breast malignant lesions. The efficacy of quantitative parameters alone and in combination with DCE-MRI, was evaluated by the receiver operating characteristic (ROC) curve. Results:Of 116 patients, 27 patients were benign and 89 patients were malignant. T 2pre, PD pre, T 2Gd, PD Gd, ?T 2, T 1ratio and T 2ratio values of malignant patients were lower than those of benign lesions ( P<0.05). Multivariate logistic regression analysis showed that T 2pre (OR=0.961,95%CI 0.926-0.993, P=0.022)and PD pre(OR=0.945, 95%CI 0.908-0.979, P=0.003) were independent predictors of breast malignant lesions. The area under ROC curve (AUC) of T 2pre+PD pre in differentiating benign and malignant breast lesions was 0.77, and the accuracy, sensitivity, specificity, positive predictive value and negative predictive value were 0.84, 0.87, 0.68, 0.92 and 0.56, respectively. The AUC of T 2pre+PD pre combined with DCE-MRI in differentiating benign and malignant breast lesions was 0.95, indicating the best diagnostic efficiency, and higher than that of DCE-MRI alone (0.92)( Z=2.20, P<0.05). Conclusions:T 2pre and PD pre values of synthetic MRI can be used as reference parameters for differentiating benign and malignant breast lesions. Synthetic MRI combined with DCE-MRI can effectively improve the diagnostic efficiency of breast malignant lesions.

A Light Adaptive Heart Rate Detection Method Based on Webcam / 中国医疗器械杂志

In order to extract the pulse wave signal of blood volume effectively in the case of uneven light, a light-adaptive heart rate detection method based on webcam was proposed. In this method, adaptive gamma transform is applied to face image sequence to eliminate the influence of illumination. The pulse wave source signal was extracted from the forehead area and the blood volume pulse wave was obtained by wavelet filtering. The heart rate is estimated by Fourier transform analysis. The Bland-Altman analysis indicates that the method used in this paper is in good agreement with the measurement results of the electronic sphygmomanometer, and the adaptive gamma transformation used in this paper eliminates the influence of light interference, and the measurement error of heart rate is significantly reduced, which is completely able to meet the requirements of daily heart rate monitoring.

Correlation between gestational age and ADC value in normal fetal brain region detected using DWI with b value as 1 000 s/mm2 / 中国医学影像技术

Objective: To investigate the relationship between ADC value of normal fetal brain region obtained on ADC image generated with DWI of b value as 1000 s/mm2 and gestational age. Methods: A retrospective analysis of 106 pregnant women who underwent routine MR and DWI examinations due to suspected fetal abnormalities with prenatal ultrasonography was performed. Among them, 28 pregnant women were in the second trimester (13-27 weeks) and 78 pregnant women were in the third trimester (≥28 weeks). All fetuses were confirmed as normal neurodevelopment with prenatal routine MR and DWI examinations, as while as postnatal follow-up. DWI was performed with b value of 1000 s/mm2. ADC maps were generated based on DWI,and ADC values of frontal lobe, parietal lobe, temporal lobe, occipital lobe white matters, basal nucleus, thalamus, cerebellar hemisphere and pons were measured. The correlation between ADC values in different fetal brain regions and gestational age was analyzed using linear regression. Results: ADC values of different parts of normal fetal brain were in order from high to low: Parietal white matter ([1.84±0.08]mm2/s), occipital white matter ([1.82±0.08]mm2/s), frontal white matter ([1.77±0.03]mm2/s), temporal white matter ([1.70±0.13]mm2/s), cerebellar hemisphere ([1.41±0.10]mm2/s), basal ganglia ([1.40±0.13]mm2/s), thalamus ([1.35±0.04]mm2/s), pons ([1.27±0.08]mm2/s). Except for frontal white matter, ADC values of the rest of the normal fetal brain regions were negatively correlated with gestational age (all P<0.05). Conclusion: Most of ADC values in normal fetal brain different regions decreased with increasing gestational age, which can reflect the developmental process of fetal brain.