<abstract><p>Bone age assessment is of great significance to genetic diagnosis and endocrine diseases. Traditional bone age diagnosis mainly relies on experienced radiologists to examine the regions of interest in hand radiography, but it is time-consuming and may even lead to a vast error between the diagnosis result and the reference. The existing computer-aided methods predict bone age based on general regions of interest but do not explore specific regions of interest in hand radiography. This paper aims to solve such problems by performing bone age prediction on the articular surface and epiphysis from hand radiography using deep convolutional neural networks. The articular surface and epiphysis datasets are established from the Radiological Society of North America (RSNA) pediatric bone age challenge, where the specific feature regions of the articular surface and epiphysis are manually segmented from hand radiography. Five convolutional neural networks, i.e., ResNet50, SENet, DenseNet-121, EfficientNet-b4, and CSPNet, are employed to improve the accuracy and efficiency of bone age diagnosis in clinical applications. Experiments show that the best-performing model can yield a mean absolute error (MAE) of 7.34 months on the proposed articular surface and epiphysis datasets, which is more accurate and fast than the radiologists. The project is available at <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://github.com/YameiDeng/BAANet/">https://github.com/YameiDeng/BAANet/</ext-link>, and the annotated dataset is also published at <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://doi.org/10.5281/zenodo.7947923">https://doi.org/10.5281/zenodo.7947923</ext-link>.</p></abstract>

Background The neuroimaging manifestations of eclampsia and preeclampsia often overlap, mainly presenting as posterior reversible encephalopathy syndrome (PRES). The purpose of this retrospective study was to compare the extent and nature of brain edema in eclampsia and preeclampsia patients with PRES based on MRI characteristics. Methods One hundred fifty women diagnosed with preeclampsia-eclampsia and undergoing cranial MRI were enrolled; 24 of these were diagnosed as having eclampsia. According to clinicoradiologic diagnosis of PRES, eligible patients were classified as having eclampsia with PRES (group E-PRES) and preeclampsia with PRES (group P-PRES). A scale on T2W FLAIR-SPIR images was established to evaluate the extent of brain edema, and the score of brain edema (SBE) of both groups was compared. In patients of the two groups who also underwent DWI sequence, the presence or absence of hyperintensity on DWI and hypointensity on ADC maps were determined to compare the nature of brain edema. Furthermore, clinical and biochemical data of the two groups were compared. Results The incidence of PRES in eclampsia patients was significantly higher than that in preeclampsia patients (87.50% vs. 46.03%, P<0.001). The SBE of all regions and typical regions in group E-PRES patients were significantly higher than those in group P-PRES patients (15.88±8.72 vs. 10.90±10.21, P=0.021; 8.52±3.87 vs. 5.01±4.19, P=0.002; respectively). The presence of hyperintensity on DWI was determined more frequently in group E-PRES patients than group P-PRES patients (71.43% vs. 32.00%, P=0.024). Age, systolic blood pressure, white blood cell count, neutrophil count and percentage of neutrophils were significantly different between the two groups (P<0.05). Conclusions Certain MRI characteristics that reflect the extent and nature of brain edema were different between eclampsia and preeclampsia patients with PRES. Additional prospective studies are still required to explore whether these MRI characteristics of brain edema may further become a potential predictor for eclamptic seizures in preeclampsia patients with PRES.

Background The endometrium and uterine junction zone often change throughout the menstrual cycle. Some pathological conditions may appear normal in uterine imaging, which will lead to missed diagnosis and misdiagnosis. Purpose To evaluate the changes in the thickness and apparent diffusion coefficient (ADC) values of the endometrium and uterine junction zone throughout the menstrual cycle in magnetic resonance imaging (MRI) of women of reproductive age. Material and Methods Data were collected from 40 healthy women of reproductive age with regular menstrual cycles from January 2017 to April 2018. They underwent four total MRI sessions during the menstrual, proliferation, and early and late secretive phases. The main MRI sequences were T2-weighted (T2W) volume isotropic turbo spin echo acquisition (VISTA) spectral attenuated inversion recovery (SPAIR) and diffusion-weighted imaging (b = 0, 600, 800, 1000 s/mm2), which were used to measure the thicknesses and ADC values of endometrium and uterine junction zone. Results First, the endometrium was thinnest during the menstrual phase and thickest in the late secretive phase. Second, the uterine junction zone was thinnest in the late secretive phase and thickest in the menstrual phase. Third, the ADC values of the endometrium were lowest in the menstrual phase and peaked in the early secretive phase. Finally, the ADC values of the uterine junction zone were lowest in the menstrual phase and peaked in the late secretive phase. Conclusion The endometrium and uterine junction zone showed cyclic changes. Radiologists should consider these changes in the thickness and ADC values when analyzing MRI images of the uterus.